Report HPC

ISBN 9788886037440
Cineca HPC Report 2021-2022
website: www.hpc.cineca.it
mail: info-hpc@cineca.it
Contents are © Cineca 2021- 2022 are released under
the Attribution-NonCommercial-NoDerivs (CC BY-NC-ND)
Creative Commons license, version 3.0.
Editorial Office
Cineca Consorzio Interuniversitario
Via Magnanelli 6/3
40033 - Casalecchio di Reno (BO) - Italy
Editorial Board
Sanzio Bassini
Gabriella Scipione
Antonella Guidazzoli
Maria Chiara Liguori
Design and Layout project
Simona Caraceni
Print
Grafiche Zanini
Via Emilia 41/e - 40011 Anzola dell’Emilia BO
Thanks to:
Fabio Affinito; Bhaskar Agrwal; Paola Alberigo; Antonio Aloisio; Claudio Arlandini; Tiziana
Bassi; Sanzio Bassini; Giovanni Bellavia; Neva Besker; Elisa Borghi; Michele Bottazzi;
Alessandro Bucci; Simona Caraceni; Salvatore Cataudella; Cinzia Caroli; Mirko Cestari;
Patrizia Coluccia; Maurizio Cremonesi; Chiara Della Casa; Daniele De Luca; Francesca
Delli Ponti; Andrew Emerson; Francesco Falciano; Daniela Galetti; Thomas Geenen; Luca
Govoni; Massimiliano Guarrasi; Antonella Guidazzoli; Francesco Iannone, Silvano Imboden;
Mariella Ippolito; Leon Kos; Maria Chiara Liguori; Luisa Mantovani-Löffler; Markus Löffler;
Marcello Maranesi; Alessandro Marani; Gian Franco Marras; Paola Nassisi; Maurizio Ortali;
Daniele Ottaviani; Giorgio Pedrazzi; Matteo Picchiani; Elda Rossi; Elisa Rossi; Gabriella
Scipione; Donatella Sforzini; Claudia Truini; Roberta Turra; Gregorio Vlad, Cinzia Zannoni,
Paolo Zuzolo.

Dear colleagues,
the previous annual report summarizing the activities of the Cineca HPC department for the
2020-21 period focused on the united response of the scientific community to the COVID
pandemic. A year later, the focus has shifted to the intense work that is facing us as early
as the next twelve months to enhance the economic resources that have been mobilized
with the Next Generation Europe action to recover from the effects of the pandemic and
to support the priorities shared by the Member States of the European Union and by the
international community.
In particular, we need to confront the ecological and energy transitions necessary to mitigate
climate change, especially development sustainability and digital innovation.
At a national level, Next Generation EU funding from the National Recovery and Resilience
Plan(PNRR)hasbeenstructuredintosixmissions:Digitization,innovation,competitiveness,
culture and tourism; Green revolution and ecological transition; Infrastructures for
sustainable mobility; Education and research; Inclusion and cohesion and Health.
Cineca takes part in the fourth mission, Education and research, which is managed by
the Ministry of University and Research (MUR) to which Cineca belongs. In this mission,
specific implementation tools were identified: key technologies, research infrastructures,
extended public-private partnerships and territorial ecosystems and Cineca is involved in
funded projects which belong to all the fields in these sectors. The first is the project for the
establishment of the National HPC, Big Data and Quantum Computing Foundation: Cineca
is one of the promoters and founders of this center which is funded by the key technologies
mechanism. HPC and its holistic ecosystem is one of the five key technologies defined by
the MUR and associated with this action is also that relating to the financing of research
infrastructures and Cineca is involved as an implementing body in some projects funded
by this method. Overall, the economic resources assigned to Cineca for the enhancement of
the national HPC infrastructure and the increase in staff for specialist support will be over
80 million euros.
Inthefieldofkeytechnologies,CinecaalsoparticipatesasafoundingmemberintheNational
Center for the Biodiversity. In actions financed with the extended partnerships mechanism,
we participate in those relating to the Growing Resilient, Inclusive and Sustainable area and
to the Translational Medicine area. As regards Ecosystems for territorial innovation, we are
involved in the project relating to the territory of the Emilia Romagna Region for the area
of Sustainability of Economic Development. Overall, the projects that see Cineca as a core
partner have been funded with approximately one billion euros, of which approximately
10% has been directly assigned to Cineca.
This confirms on the one hand the centrality of the roles of HPC, ML and AI and of big data
processing as enabling assets for advanced research and technological innovation, on the
other hand it is recognition of the quality and contribution that Cineca can provide.
Naturally, the activity carried out over the last twelve months has been particularly
characterized by the installation of the pre-exascale system of the Leonardo project, which
will start production at the beginning of 2023. The Leonardo system is the most important
project for the next period and we are already committed to upgrading the computing system
to extend the general purpose partition with the integration of a partition configured with
microprocessors based on HBM technology and the booster partition configured with the
latest accelerators. Furthermore, the integration of a quantum computing system is already
foreseen, and this will complete the tier0 infrastructure for the scale-out computing.
The evolution of the computing systems and the national HPC infrastructure, integrated and
federated into the European HPC system must also include an investment in the staff of the
high-level support team. To achieve this there will be an action which will see an increase
in the next six months of at least fifty new structured figures in the department dedicated
to this activity, to the national and European scientific communities, and to the flagship
projects both at international levels such as Destination Earth and Copernicus, and at a
national level, particularly the National Center for HPC, Big Data and Quantum Computing.

2
4
7
8
14
15
17
18
22
23
26
27
33
2021 IN NUMBERS
HPC Infrastructures
Leonardo
ICEI
Users’ statistics
Resource allocation: projects, agreements, collaborations
PRACE
Iscra
EUROFusion
Agreements
Training HPC
Staff
Events
HPC H-index
SCIENTIFIC OVERVIEW
LIFE SCIENCES
ORCHESTRA
ARNO
EARTH AND ENVIRONMENTAL SCIENCES
SEBASTIEN
Digital technology enablement for Destination
Earth
36
38
39
41
42
43
44

H
b
G

BIG DATA AND ARTIFICIAL INTELLIGENCE...
Ocean, Supercomputers and AI in action
Day-aheadhailpredictiontoprevent
lossesinagriculture
MATERIAL SCIENCES
MaX: HPC and HTC meet for Material
Sciences
QUANTUM COMPUTING .
Cineca quantum computing perspective
HUMANITIES AND SOCIAL SCIENCES, DIGITAL ECOSYSTEMS
The great cycle: the immersive experience of
TUBKE Monumental
DARE: a digital environment for urban
regeneration
DARE: IMC platform and the management of 3D
digital objects
COMPUTING PROVISIONING SERVICES
INDUSTRIES....
Bringing HPC-fueled innovation to the
EU industry
EUROCC
NATIONAL RECOVERY AND RESILIENCE PLAN
High Performance Computing projects supported
by the National Recovery and Resilience Plan – Next
Generation EU
46
47
49
51
52
54
55
58
59
62
64
66
67
68
72
73
74

1
Overview
2017
Cineca HPC is the main
Center for scientific
computing in Italy. It
runs a large computing
infrastructure and
make it available to
Italian and European
Researchers, as well
as to important
Italian Companies
within a programme
for supporting
national industrial
competitiveness.
Here we present data
about usage and users,
projects, events and
classes offered in
the field of High
Performance Computing.
At the end of this
section we present
our staff, the true
resource that makes
everything work.

2
HPC infrastructures
Elda Rossi, Massimiliano Guarrasi
Cineca
Starting from March 2021, the HPC system
Galileo was taken off the production
and substituted with a new and more
performant structure called Galileo 100,
co-funded by the European ICEI (Interactive
Computing e-Infrastructure) project and
made available on August 2021. Engineered
by DELL, this new infrastructure is based
on PowerEdge processors and NVIDIA GPU
accelerators and features 554 computing
nodes, subdivided into 340 standard
nodes, also called “thin nodes”, 180 data
processing nodes, also called “fat nodes”
and 34 GPU nodes interconnected by a
100Gbs Infiniband interconnection and
dedicated to visualization purposes, thus
also called “viz nodes”. Beyond these nodes,
the infrastructure is equipped with 77
computing servers OpenStack for Cloud
Computing since the Cloud Infrastructure,
hosted on Meucci, was enhanced and moved
onto this partition. The whole architecture
is interconnected by a Mellanox Infiniband
100GbE internal network and it is supported
by a 20 PB active storage, accessible from
both the cloud and compute partition of
Galileo 100, 1PB Ceph storage (full NVMe/
SSD) for the cloud partition and 720 TB
fast storage. Galileo100 is mainly used for
special High-End projects for the technical
and industrial HPC computing, as well as
for dedicated Climate and Weather projects,
reachingapeakperformanceof3.53TFlop/s
per single node and 2PFlop/s in total.
The Tier-0 cluster Marconi100 and
Marconi-A3 remained unchanged in the
2021.Furthermore, on January 2021, a new
specialised system for AI, based on NVIDIA
DGX, has been made available for users. It
is composed by 3 nodes and features Dual
AMD ROME 7742 processors, accelerated by
8x NVIDIA A100 GPUs per node.
Finally, in addition to aforementioned
systems, other infrastructures are present
and dedicated to special uses or to specific
communities.
These includes:
GATEWAY: 28 nodes (Skylake Intel Xeon
8160) with 28 cores each, connected with
a OmniPath network. It is a dedicated
interactive cluster for EUROfusion.
CNAF: 288 Broadwel nodes (Intel Xeon E5-
2697) connected with a custom internal
network and linked with a high performance
link with the Bologna INFN computing
room. The cluster is dedicated to INFN
users for WLCG program.

3
CPU Total nodes/
cores
Memory/
node
GALILEO 100
(DELL)
Dell Dual Socket PowerEdge +
NVIDIA Volta 100
554 nodes
26,592 cores,
2x554 GPUs
348 GB +
3.0TB Opta-
ne on 180 fat
nodes
MARCONI 100
(IBM)
IBM Power AC922 + NVIDIA Volta
100
980 nodes
31.360 cores,
4x980 GPUs
256 GB
MARCONI A3 –
SKL (LENOVO)
Intel SkyLake 2x Intel Xeon 8160
@2.1GHz 24 cores each
3.216 nodes
154.368
cores
192 GB
Supercomputer Resources (Core Hours) Performance capacity (Exaflops)
Galileo 96.771.456 14.516
Marconi A3 1.340.490.240 321.718
Marconi 100 274.589.184 988.521
Galileo100 58.715.136 15.545
Total 1.770.566.016 1.340.299

4
Leonardo
Mirko Cestari
Cineca
The rapid growth of the amount of data
coming from diverse sources, including
numerical models and sensors, requires
more and more computing resources,
able to process data in a timely fashion.
Furthermore, the launch of new leading
Italian and European initiatives, which
face increasingly advanced computational
challenges, are based on the availability of
these technologies. Leonardo will be one
of the most powerful high-performance
computing (HPC) systems worldwide,
capable of executing more than 250
Petaflops, i.e., more than 250 quadrillion
(or 1015) operations per second. HPC
systems of Leonardo class are key
enabling technologies, aiming to raise the
competitiveness of Italian and European
research and paving the way to the
forthcomingexafloperainEurope.Leonardo
can be considered a significant step forward
in promoting European research in the field
of computational sciences. The goal is to
strengthen the European presence in high-
performance computing, a strategic asset
aiming to advance the technological growth
of the member states of the Union.
Cineca has a proven history of providing the
most powerful supercomputers in the world.
With the support of national and regional
government, Cineca was designated by
the EuroHPC initiative to host and operate
Leonardo, one of the three recently
procured precursors of Exascale systems.
Leonardo will be installed at the newly
built data center located in the Bologna
Big Data Technopole beside the ECMWF
premises. The Technopole can therefore be
considered one of the main European hubs
for computing and data processing.
Leonardo features almost 5000 computing
servers and over 100 PB of storage, all
mounted on 155 Atos Sequana X2000
racks. Computing nodes are powered by
state-of-the-art technology. The accelerated
partition is constituted of 3456 nodes each
equipped with 4 customized NVIDIA A100
GPUs. The CPU partition features next-
gen Intel Sapphire Rapids processors and
512 GB of DDR5 RAM. All 5000 computing
nodes are interconnected via an NVIDIA
HDR200 based network with a Dragonfly+
topology to achieve maximum throughput
and minimal latency. The storage is based
on two DDN appliances targeting capacity
and high IOPS.
An important consideration is the
energy efficiency of the system and its
integration in the data center. The IT power
consumption of the system is expected
to reach over 6 MW, peaking at 9MW
during burst runs. It is therefore especially
important to reduce at the minimum any
source of energy consumption other than
IT. All the computing part of the system is
under direct liquid cooling (cold plate) able
to extract over 95% of the dissipated heat.
The cooling water in the racks operates at
a temperature range of 36-46 degrees and
is provided through 4 independent dry
cooling systems. This is the main reason
for the estimated power usage effectiveness
of Leonardo in the order of 1.1, classifying
it as an extremely energy efficient system.
Leonardo is expected to be operational from
Q1 2023.

5
Figure 1: high-level logical overview of Leonardo supercomputer archi-
tecture
Figure 2: Leonardo Cineca

6
Figure 2: ..
Figure 3: QR Video teaser Leonardo

7
ICEI
Interactive Computing E-Infrastructure
Mirko Cestari
Cineca
Interactive Computing E-Infrastructure
– in short ICEI – is a project aiming to
provide a set of composable services to be
used to build the Human Brain Project and
other scientific community driven digital
platforms. In other words, the partners of
the project (Cineca, BSC, CEA, CSCS and
Juelich) have agreed on a framework to
provide computing and storage services
with common features. For example: cloud
services based on the Openstack suite, or
object storage for archival data repository
services.
To achieve this goal the project partners
have taken responsibility for a series of
procurement procedures for R&D services
and hardware. Cineca has taken the leading
roles in the provisioning of two R&D services
and one hardware system. The latter has
led to the procurement and installation
of Galileo100. This system started the
operations in July 2021 and production in
the following September. Given the goal of
the project, Galileo100 was the first Cineca
HPC system providing multiple services
thorough a convergent (tightly integrated)
infrastructure. In fact, 5 main services,
each powered by different hardware
infrastructure, can be identified:
• Scalable computing (SCC). Galileo100
is provided with 340 “thin” nodes to host
small-to-mid size parallel jobs. Each node
is equipped with: 2x CPU 8260 Intel Cascade
Lake 24 cores, 384 GB RAM DDR4 2933MT/s
and 480 GB SSD local disk
• Interactive computing (IAC). Galileo100
offers on top of the same configuration of
SCC, 34 nodes with 2 NVIDIA V100 GPUs
and 180 nodes with 3 TB of Intel persistent
memory. This partition is particularly suited
for use cases that require large memory
servers and intensive I/O workloads, as well
as GPU resources.
•Cloudservices(Cloud).Galielo100provides
77 computing servers configured as the SCC
partition but with double the size of RAM
memory (768 instead of 384 GB). To service
the Cloud, the system features 12 OpenStack
storage nodes providing a total of 720 TB of
dedicated CEPH storage full flash (NVMe/
SSD) for high IOPS. This storage will host VM
root disks, disk volumes and VM snapshots.
• Active data repository (ACD). Galileo100
features over 20 PB of online storage
with adequate performance for hosting
simulation data
• Archival data repository (ARD).
Galileo100 offers object storage with S3 API
functionalities for long term data storing.
Lastly, Cineca took the lead to procure on
behalf of the project partners an Interactive
computing service. This will be based on the
JupyterHub application suite adapted by
E4 and co-designed for the specifics of the
ICEI systems. The result of this R&D project
will be installed on the Galileo100 system
exploiting the IAC partition and ported on
the future systems provided by Cineca.
Fenix has received
funding from the Euro-
pean Union’s Horizon
2020 research and
innovation program-
me through the ICEI
project under the grant
agreement No. 800858.

8
A
CNR
S
c
A
d
v
P
I
S
A
R
O
M
A
F
O
R
O
R
O
M
A
L
A
S
A
P
I
E
N
Z
A
R
O
M
A
T
O
R
V
E
R
G
A
T
A
RO
M
A
TRE
S.RAFFAELE
SALENTO
SALERNO
UNISANNIO
SASSARI
SIENA
TERAMO
TORINO
TRENTO
TRIESTE
TUSCIA
UDINE
UNINETTUNO
URBINO
VENEZIA
VERONA
GRAN
SASSO
LUCCA
O
T
H
E
R
PAVIA
P
I
S
A
S
.
A
N
N
A
S
I
S
S
A
12
1
2
4
5
1
0
1
7
6
79
7
3
7
17
1
1
7
0
48
32
108
59
12
2
2
7
5
219
4
1
0
1
3
5
9
3
3
9
1
8
4
0
OTHER I.A.
On our HPC infrastructure, the active users at
the end of 2021are 4295, 468 more than the
year before. A large part of the users are males
(78%),43% of them are aged between 31 and
45 years old while 35% are less than 35 years
old.
73% of the total number of users works for an
Italian institution, clustered in the Lombardia
region (21%), Emilia Romagna region (18%)
and Lazio and Friuli (12-15% each), while
going to city level, Milan, Rome, Bologna and
Trieste, in the order, have a user’s percentage
going from 18% to 11%.
The more represented foreign countries are
Germany with a 5% and United Kingdom,
France, Spain and China with 3%, mainly
working for universities and public or non-
profit organizations.
Our computational resources are used by
scientists within all scientific disciplines,
the most represented three are Chemistry
and Material Science (26%), Engineering
and Computational Fluid Dynamics (15%),
Condensed Matter Physics (13%).
Resource
Users Statistics
Elda Rossi, Massimiliano Guarrasi
Cineca
2
0
2
1
2
0
2
0
2
0
1
9
2
0
1
8
2
0
1
7
2
0
1
6
2
0
1
5
2
0
1
4
2
0
1
3
0
1000
900
800
700
600
500
400
300
200
100
257
257
257
131
131
131 146
146
146
339
339
339 352
352
352 352
352
352
541
541
541
880
880
880
960
960
960
Figure 4: new HPC
users per year

9
105
OTHER
TORINO
4
Italian
I
n
t
e
r
n
a
t
i
o
n
a
l
PublicResearchInstitutes O
t
h
e
r
R
e
s
e
a
r
c
h
I
n
s
t
i
t
u
t
e
Polytechnic
Academia
c
h
o
ol of
v
a
n
c
e
d
Studies
P
r
i
v
a
t
e
U
n
i
v
e
r
s
i
t
i
e
s
P
u
b
l
i
c
o
r
n
o
n
-
p
r
o
f
i
t
P
r
i
v
a
t
e
In
te
rn
a
l
Ot
he
r
O
th
e
r
in
te
r
n
a
ti
o
n
a
l
in
s
ti
tu
te
s
ARE
A
E
N
E
A
II
T
IN
A
F
I
N
F
N
I
N
G
V
I
N
R
I
M
O
G
S
O
T
H
E
R
A
R
P
A
C
I
M
A
O
THER
T
E
L
E
T
H
O
N
BARI
MILANO
BARI
BASILICATA
BERGAMO
BOCCONI
BOLOGNA
BOLZANO
BRESCIA
CAGLIARI
CALABRIA
CAMERINO
UCBM
CASSINO
CATANIA
CATTOLICA
CUSANO
ENNA
KORE
FERRARA
F
I
R
E
N
Z
E
G
.
D
'
A
N
N
U
N
Z
I
O
G
E
N
O
V
A
H
U
M
A
N
I
T
A
S
I
N
S
U
B
R
I
A
L
'
A
Q
U
I
L
A
M
A
G
N
A
G
R
A
E
C
I
A
M
E
S
S
I
N
A
M
IL
A
N
O
M
IL
A
N
O
-B
IC
O
C
C
A
M
O
D
E
N
A
e
R
.E
.
NAPO
LI
-PART
HENO
PE
NAPO
LI
-
SECO
NDA
NA
PO
LI
FE
DE
RI
CO
II
P
A
D
O
V
A
P
A
L
E
R
M
O
P
A
R
M
A
P
A
V
I
A
P
E
R
U
G
I
A
P
I
E
M
O
N
T
E
A
O
t
h
e
r
I
t
a
l
i
a
n
I
n
s
t
i
t
u
t
i
o
n
s
1
1
6
9
7
0
5
7
1
2
2
5
0
3
2
1
6
3
12
13
227
10
0
17
38
166
1
34
19
24
3
2
1
7
3
0
0
39
4
2
6
2
7
2
6
3
0
3
7
9
2
1
2
9
7
0
0
6
3
9
8
4
1
7
2
7
3
3
2
8
3
4
3
3
0
5
5
5
4
124
106
Figure 5: classification of
users’ institutions

10
Figure 6: geographical distrubution of international users

12
Figure 7: geographical distribution of Italian users

13
Figure 8: distribution of users by disciplines
Other
Computational Fluidodynamics
Condensed Matter
Computational Chemistry
Astrophysics and Plasma
Computational Engineering
Nuclear Fusion
Life Science - Computational Biology
Earth and Climate Sciences
Life Science - Bio Informatics
Particle Physics
Mathematics
AI and Machine Learning
Humanities and Social Sciences
522
522
518
518
485
485
460
460
426
426
421
421
394
394
268
268
251
251
209
209
175
175
79
79
55
55
32
32

14
A very significant part of the allocations on Cineca’s system are based on a peer review
selection in order to ensure the highest scientific value of the selected projects. In fact, Italian
and European researchers take advantage of the HPC clusters in Cineca for performing
their computational research, applying for “computational projects” which allow for access
to the HPC facilities. PRACE and ISCRA programs allocate the majority of the resources on
Marconi100 and Galileo100 and they are both based on the peer-review mechanism at the
European and national level respectively . On the other hand, Institutions which want to
use HPC infrastructures can sign an “Agreements” with Cineca as well as the “Industrial”
users. Generally, a project is characterized by a starting date, an end date, a science domain,
a budget (in terms of core-h/GPU hours) and a PI (Principal Investigator) with several
collaborators. Along with these wide classes of projects, there are dedicated partitions on
MarconiA3andGalileo100forweatherandbiogeochemicalforecastactivitiesandindustrial
projects while some special partitions of Marconi100 and Marconi A3 are dedicated to the
EUROfusion consortium. For these partitions, the allocation is made autonomously by the
dedicated committee. Considering the whole Cineca HPC infrastructure during 2022 542
million core-h have been made available out of a total allocation of 657 million core hours.
This means that the systems have been overallocated at 121.2% - this is due to the huge
request for computing resources by the scientific community The most significant part of
the resources in terms of core hours has been dedicated to agreements (63%), as well as
25% to PRACE, 25% to ISCRAand 2% to industrial projects.
Resource allocation: projects, agreements,
collaborations
Elda Rossi
Cineca
Prace
131,952,983
Prace
131,952,983
Projects
342,702,081
Projects
342,702,081
Iscra
134,779,743
Iscra
134,779,743
Other
2,019,122
Internal
576,691
ICEI/HBP
10,061,741
Contracts
10,596,471 European Projects
23,181,395
Trial
649,471
Figure 9: HPC allocation (Core hours) by project type

15
During 2021, two PRACE calls have been
awarded on Marconi100: call 22 and call 23.
For the regular calls a total of 23 projects
have been allocated on Marconi100; 11
of those have been assigned to Italian PIs.
According to the PRACE-2 agreement, as
in the previous calls, 5% of the PRACE
dedicated computational resources have
been awarded to Center of Excellence (23
projects). The two calls distributed a total
of 15.104.495 GPU hours (corresponding
to 120.835.945 Local Core Hours or
1.329.195.000 PRACE Cumulative Core
Hours), as reported in the following table.
PRACE
Partnership for Advanced Computing in Europe
Massimiliano Guarrasi, Elda Rossi, Gabriella Scipione
Cineca
Accepted Call 22
1 April, 2021 to Thursday, 31
March, 2022
Call 23
1 October, 2021 to Friday, 30
September, 2022
Regular call 12 11
Reserved to CoE 12 11
Total projects for
call
24 22
Total accepted
projects
46
Marconi 100 Node
Hours / Local Core
Hours
59.965.909 60.870.036

17
ISCRA (Italian Supercomputing Resource
Allocation) gives computational resources
and technical support to researchers
affiliated to Italian institutions, in order to
support the Italian research. It provides
HPC resources through Class C projects (for
code testing and pre-production) and Class
B projects (full production). IscraC projects
are evaluated and selected on a continuous
basis, IscraB projects twice a year, on the
basis of a scientific peer review procedure.
In 2021 Cineca provided to the Italian
research community 107M core-h on
Galileo100 and Marconi100. In particular
two IscraB calls were launched as reported
inthetablebelow(call23and24).Since2021
in addition to HPC resources also Cloud,
DGX and Quantum Simulator resources are
available through the ISCRA programme.
Iscra
Italian SuperComputing Resource Allocation
Paola Alberigo
Cineca
call 23B
19 February, 2021 to Thursday,
19 May, 2022
call 24B
25 October, 2021 to Tuesday,
25 April, 2023
Accepted 34 40
Rejected 36 31
Total number sub-
mitted projects
141
Total accepted
projects
74
Call 23B Call 24B Local Core Hours
DGX 40.000 4.664.480 40.000
GALILEO100 2.104.998 31.151.760 33.256.758
MARCONI100 37.354.272 32.981.152 70.335.424
MARCONI A2 4.000.000
Total 107.632.182

18
Supercomputing is an important aspect
of nuclear fusion research as it plays a
crucial role in the modelling of the plasma
and materials, validating the experimental
results of fusion devices and designing the
next-generation fusion reactor.
The implementation of the EUROfusion
Consortium programme has benefited in
2021 from dedicated high performance
computing resources provided through
the HPC Marconi-Fusion and the Gateway
operated by Cineca in Bologna. As foreseen
in the Project Implementing Agreement
of Marconi-Fusion, the operation of the
systems has been successfully monitored
and supervised by the HPC Project
Committee,withthesupportoftheOperation
Committee and the Ticket Committee.
Availabilityoftheresources(mostimportant
key performance indicator) was good and
compliant with the Service Level Agreement
for all partitions throughout the year with
about ≈21 Mnode-h of usage for about ≈24
Mnode-h of availability of Marconi Fusion
Skylake partition.
The EUROfusion computing projects
running on HPC systems at Cineca are
selected by the Allocation Committee on the
base of their relevant contributions to the
following topics:
• Plasma turbulence and related transport
processes
• Fast particle physics
• Non-Linear and/or Extended MHD
• Edge physics
• Heating and Current Drive
• Reactor Materials
• Reactor Technology.
The following table shows the computing
time allocated in the six annual cycles since
the HPC for Fusion services were provided
at Cineca.
Plasma Physics
The plasma physic turbulence is still the
topic requiring a large computing time
as a key issue for future burning plasma
experiments, such as ITER. Multiphysics
HPC simulations are undergoing to
Verification and Validation phases by
means integrated plasma models for the
Energetic Particles (EP) transport due to
mesoscale Alfvénic instabilities primarily
excited by EPs and EP coupling with
microturbulence and macroscopic MHD
modes mostly driven by thermal plasmas
(Taimourzadeh,2019). Initial V&V studies
of the linear gyrokinetic simulations of
Reversed Shear Alfvén Eigenmodes (RSAEs)
in tokamaks are still carrying out using a
gyrokinetic particle codes. Good agreement
in RSAE frequency, growth rate, and mode
structure have been obtained among these
simulations, and between simulation results
and experimental measurements. The
successful linear V&V lends some degree
of confidence to nonlinear gyrokinetic
simulations that provide new kinetic
insights on nonlinear Alfvén eigenmode
dynamics and EP transport, and help the
construction of reduced EP transport
models which are needed for fast parameter
EUROFusion
Massimiliano Guarrasi, Elda Rossi
Cineca
Francesco Iannone, Gregorio Vlad
ENEA
Leon Kos
University of Ljubljana

19
scans, shot-to-shot analysis, and optimization of ITER experiments.
One of the major challenges in magnetic confinement thermonuclear fusion research
concerns the confinement of the energetic particles (EPs) produced by fusion reactions and/
or by additional heating systems. In such experiments, EPs can resonantly interact with the
shear Alfvén waves. In the frame of the EUROfusion 2019–2020 Enabling Research project
‘multi-scale energetic particle transport in fusion devices’ (MET), a detailed benchmark
activity has been undertaken among few of the state-of-the-art codes available to study the
self-consistent interaction of an EP population with the shear Alfvén waves. In Vlad,2021
linear studies of EP driven modes with toroidal mode number n = 1 are presented, in real
magnetic equilibria and in regimes of interest for the forthcoming generation devices (e.g.
ITER, JT-60SA, DTT). The codes considered are HYMAGYC, developed in ENEA Frascati,
MEGA, and ORB5, the first two being hybrid MHD-gyrokinetic codes (bulk plasma is
represented by MHD equations, while the EP species is treated using the gyrokinetic
formalism), the third being a global electromagnetic gyrokinetic code.
The following figure shows the results of the three codes for the two reference cases (peaked
on-axis and off-axis EP density profiles): in both cases, for all the three codes considered,
a mode driven by the EPs is observed. As an example, let’s consider, first, the results for
the peaked on-axis EP density profile (left column): after an initial transient phase of the
simulations, both HYMAGYC and MEGA observe, as the most unstable one, a mode located
around s ≈ 0.4, with a dominant m = 2 poloidal Fourier component for the electrostatic
field φ, and located around ω ≈ −130 kHz in frequency, i.e. with a frequency slightly lower
(in absolute value) than the lower shear Alfvén continuous spectrum. These observations,
together with the fact that its radial location is quite close to the minimum q value, suggest
that this mode is a so-called RSAE. On the other side, ORB5, for the same parameters,
observes an externally localized TAE (Toroidal Alfvén Eigenmode) with frequency ω
≈ −200 kHz . This discrepancy between HYMAGYC and MEGA on one side, and ORB5 on
the other, has been reconciled by considering the dependence of the EP driven modes on
the EP density: indeed, all the three codes observe, for low EP density values, a dominant
externally localized TAE; this mode, as the EP density is raised, becomes, eventually, sub
dominant to a stronger RSAE. The occurrence of the ‘cross-over’ between RSAE and TAE
depends strongly on the different damping experienced by the codes, which reflects in
considerable variations of the net growth rates observed for this particular case (peaked on-
axis EP density profile). The results for the peaked-off axis EP density profile (right column),
on the contrary, show a remarkable agreement among the three codes, obtaining all of them
a TAE localized close to the magnetic axis, where the EP density profile is characterized by
positive radial gradient.

20
Figure 10: comparison between the results obtained for HYMGYC (top row), MEGA (centre row), ORB5 (bottom row), for the
two ASDEX tokamak nominal cases: peaked on-axis EP density profile (left column) nh0/ni0 = 0.261, Th = 0.093 MeV, and
peaked off-axis EP density profile (right column) nh0/ni0 = 0.0274, Th = 0.093 MeV. The radial profiles for different
poloidal Fourier components of the e.s. potential φ≈(s), and its power frequency spectra are plotted for each case
(linear color scale).
The HYMAGYC code, with a careful analysis of the frequency power spectra, is able to
recognize also a sub-dominant, more external mode, characterized by a negative frequency
ω ≈ −200 kHz, and located within the toroidal gap, at the radial position s ≈ 0.7 as depicted
in the following figure.
Figure 11: HYMAGYC: peaked off-axis EP density profile, with nh0/ni0 = 0.0067: e.s. field structure φ(R, Z) in the poloidal plane ϕ = 0 (left), radial profiles
of the Fourier components φm,n(s)| (centre), and frequency power spectra φ(s,ω)2
(right). The dominant core localized TAE (s ≤ 0.4), with preeminent m
= 2 poloidal Fourier component is observed in the central region, with a positive real frequency; in the external region (s ≥0.4) it is also present the sub-
dominant external TAE, with a preeminent m = 3 poloidal Fourier component and a negative real frequency (i.e. with opposite sign w.r.t. the dominant
internal mode)
Reactor Technology
The computing resources allocated in reactor technology has grown on the last allocation
cycles. The neutronics analyses performed with Monte Carlo codes are fundamental in
nuclear reactor design. Detailed neutronics analyses have been performed with MCNP5
Monte Carlo code aimed at evaluating the DEMO divertor nuclear responses to provide
inputs for thermo-hydraulic and thermo-mechanical analyses and to assess its shielding
performances. A heterogeneous neutronics model of DEMO divertor has been developed
and integrated in the DEMO MCNP model with a Water Cooled Lithium Lead (WCLL) blanket.
The model is shown in the figure below. (Figure 12)
A
B
Neutron flux, nuclear heating, cumulated damage and helium production have been
calculated in all divertor components (plasma facing components, cassette body, liner,

21
reflector plates and supports) as well as on the fixation systems, rails, in-vessel coil and
vacuum vessel. Three-dimensional distributions of a) Irradiation damage (dpa/FPY) and b)
helium production in divertor EUROFER are shown in the plots below. The results of these
analyses provided important outcomes for the optimization of the design of the DEMO
divertor.
Integrated Tokamak Modelling
ENEA and Cineca jointly provide to manage the Gateway infrastructure as platform for code
development and data analysis for tokamak experiments by means an integrated modelling
approach. Over the years, ITER has developed The Integrated Modelling and Analysis Suite
(IMAS): a scientific software infrastructure which orchestrates the collective development
and execution of integrated plasma codes and plasma applications describing fusion
operations in tokamak experiments such as ITER. Currently, the IMAS suite is available on
ITER HPC, EUROfusion Gateway HPC at Cineca and on the HPC infrastructures of some of
the ITER collaborators (JET, etc...). Plug-in for open source data analysis and visualization
are provided to interface IMAS in several models including plasma equilibrium and Scrape
Off Layer in operative conditions of a tokamak as shown in the following figure. (Figure 13)
Recent papers:
S. Taimourzadeh et al 2019 Nucl. Fusion 59 066006. Verification and validation of integrated simulation of energetic particles in fusion plasmas I: linear
simulations.
G. Vlad et al 2021 Nucl. Fusion 61 116026. A linear benchmark between HYMAGYC, MEGA and ORB5 codes using the NLED-AUG test case to study
Alfvénic modes driven by energetic particles.
Gutierrez-Milla, A., Zok, T., Owsiak, M., Plóciennik, M., & Mantsinen, M.J. (2021). Progress in the transferability of fusion workflows across HPC systems.
Plasma Physics and Controlled Fusion, 63.
J.H. You, G. Mazzone, E. Visca, H. Greuner, M. Fursdon, Y. Addab, C. Bachmann, T. Barrett, U. Bonavolontà, B. Böswirth, F.M. Castrovinci, C. Carelli, D.
Coccorese, R. Coppola, F. Crescenzi, G. Di Gironimo, P.A. Di Maio, G. Di Mambro, F. Domptail, D. Dongiovanni, G. Dose, D. Flammini, L. Forest, P. Frosi,
F. Gallay, B.E. Ghidersa, C. Harrington, K. Hunger, V. Imbriani, M. Li, A. Lukenskas, A. Maffucci, N. Mantel, D. Marzullo, T. Minniti, A.V. Müller, S. Noce,
M.T. Porfiri, A. Quartararo, M. Richou, S. Roccella, D. Terentyev, A. Tincani, E. Vallone, S. Ventre, R. Villari, F. Villone, C. Vorpahl, K. Zhang, Divertor of the
European DEMO: Engineering and technologies for power exhaust, Fusion Engineering and Design,Volume 175,2022, 113010, ISSN 0920-3796, https://
doi.org/10.1016/j.fusengdes.2022.113010
Dejan Penko, Leon Kos, Xavier Bonnin, Simon D. Pinches, EUROfusion IM Team. Post-processing for ITER Scrape-Off Layer Plasma Simulations (SOLPS-
ITER) in IMAS Framework. Journal of Fusion Energy (2020) 39:202–211.

22
Cineca computing platforms may be also
accessed by agreements for collaborations.
Several Universities and Research
Institutions decided to begin or continue
to support their researchers through
agreements that reserve an amount of
computing and storage resources without
the need of submitting for selection a project
proposal. These agreements grant access
to resources from one to a few years and
are particularly suited for training and test
projects as well as experimental activities.
The research and academic institutions
with active agreements in 2021 are listed
below.
A total of 356 Mcore-h have been granted
to agreement projects. The largest piece
of allocation has been reserved by INFN
with almost 144 Mcore-h, behind SISSA
with 74 Mcore-h and ICTP that booked a
total of 45 Mcore-h. Milano Bicocca among
the other Universities reserved the largest
slice with 38 Mcore-h. INAF also holds an
important agreement with 27 Mcore-h in
2021. The institutions in the field of Life
Science research all together sum up almost
4 Mcore-h grant. The other Universities and
Polytechnics sum up a global 14 Mcore-h.
During year 2021 more than 20 agreements
with Universities and Research institutions
were active.
Agreements
Maurizio Cremonesi
Cineca
Institution
Award resources
(Mcore-h)
INFN (Istituto Nazionale Fisica Nucleare) 143.65
SISSA (Scuola Internazionale Superiore di Studi Avanzati) 73.73
ICTP (international Centre for Theoretical Physics) 45.42
Università degli Studi di Milano Bicocca 38.17
INAF (Istituto Nazionale di Astrofisica) 29.25
OGS (Istituto Nazionale di Oceanografia e Geofisica Sperimentale) 4.28
Regione Emilia-Romagna 3.33
Università degli Studi di Milano 3.21
Università degli Studi di Trieste 3.17
Università degli Studi dell'Aquila 2.50
Politecnico di Milano 2.45
Elixir (Distributed Infrastructure for Biological Data) 2.25
Università degli Studi di Genova 1.22
Istituto G. Ronzoni 1.19
Telethon 0.99
IIT (Istituto Italiano di Tecnologia) 0.92
Università di Bologna 0.84
OspPBG (Ospedale Pediatrico Bambin Gesù) 0.71
Università degli Studi di Brescia 0.49
Università degli Studi di Bergamo 0.26
INMI (Istituto Nazionale Malattie Infettive Spallanzani) 0.25
Politecnico di Torino 0.21

23
Training has always been a hallmark of
the support activities carried on by Cineca
for the Italian research community and
users. Starting from 2012 it was extended
to European researchers, with Cineca being
recognised as a PRACE Training Center
(PTC) in HPC.
In 2021 the emergency for the pandemic
COVID-19 was not ended so we continued to
hold courses online
The teaching platform was again Microsoft
Teams and this was used as exclusive
platform for educational events in 2021 .
In 2020 the pandemic emergency came so
unexpectedly so it took some time to convert
live events in online events.
In the pre-pandemic times, the courses were
spread over three Cineca sites (Bologna,
Milan, Rome) so for almost each courses we
had three editions.
In 2021 teachers had more time to prepare
the teaching material in the online mode
so we were able to increased the number
of students accepted. Furthermore we
increased the number of students accepted,
opening the course to the largest number of
people admitting some “official students”
with full access to theory and practicals
and the rest in the waiting list as Auditors
allowing them to access to the whole course
excluding just from the support during
practicals.
Furthermore that way of teaching allowed
easily the recording of the sessions. They
were published on the courses web site,
building a rich library of teaching videos,
many of them held in English language,
making them available for the people
worldwide.
The forcing in this new way of teaching
increased our competence in terms
of organization and in using learning
platforms.
Students expressed in many fashions
the fact that we fully reached our goals in
converting process and organization.
Many comments in the surveys expressed
their likelihood in following the course being
at home, not having to plan their travels,
with no expenses and with the possibilities
to join the class at the very last minute.
During 2021 we reached these numbers: 16
online courses, and 4 schools. More than 50
people in the department were involved in
teaching, for a total of 84 days of lessons and
highly specialised training. In total, more
than 725 Italian and European researchers
tookadvantageofourtrainingprogram,with
high satisfaction levels (8,8/10), as shown
by the surveys we regularly administer.
In addition to the Cineca HPC Department
classes, our experts were involved in
partnership with external societies or
universities, giving their support in
organization or teaching in events like
Masters and Doctorate schools, workshops,
tutorials about HPC and Big data topics.
Other important initiatives: Summer of
HPC (a PRACE initiative offering young
students from all over Europe to work with
HPC teams); European Researchers Night
(Cineca is the coordinator of Society, a
European project organising the Night in
Bologna and surroundings in 2018-2019);
HPC-Europa3 (we are the coordinator of
the project offering transnational access to
researchers visiting HPC centers).
The Traning activity is regulated by two
specific procedures that refers to the ISO
standard: ISO 27001:2013 “Information
Technology - Security techniques -
Information security management system”.
Training HPC
Francesco Falciano
Cineca

25
2022 Plan
Due to the pandemic, the training events in Cineca until December 2021 were all scheduled
ONLINE since, even with controlled access, the events are assumed to be very risky.
In these first days of 2022, due to the large vaccination program, the curve of pandemic is
rapidly decreasing putting some hope for getting back to the in classroom events.
These in person events have the big advantage to allow people to connect and possibly to
establish new job-connection and new collaborative projects.
On the other hand, we notice a strong increase of the number of students interested to
follow courses, maybe for the big advantage of the possibility to follow the lessons remotely
without arrange their travel and accomodations.
In agreement with the teaching staff, in 2022 at least some courses will back in the
classrooms, but thanks also to the big experience made during these two years, we will go on
planning ONLINE events and some hybrid events for giving people the best of both worlds.
Our effort is constantly focused in increasing the number of courses and schools developed
by Cineca HPC Department, teaching the most new technologies available and trying to
admit the largest number of students worldwide is possible. For doing this we will try to
admit as much students as possible, giving them access to tutors support and, for each
course, a reasonable numbers of auditors giving them access at least to an email support
service. For some courses we will open online forums so the students will be able to discuss
and support each others with the tutoring of our experts. We will try to record and put online
large part of the lessons, tentatively in English language, making them feasible.
Furthermore we are experimenting new ways of interaction between people remotely
connected like closed rooms of discussions with few students.
Figure 14: numbers in training

26
STAFF
In the HPC department of Cineca work about
98 people, distributed among the three
premises: 77 in Bologna, 7 in Milan and 14
in Rome.
Several new colleagues were enrolled in
2021 and 2022: we wish them a very fruitful
and interesting job.
Andrea Giunchi
I completed my bachelor’s, master’s and Ph.D. degrees
in Industrial Chemistry at University of Bologna. I
worked in the field of Computational Chemistry with
particular emphasis on periodic DFT calculations
based on plane waves applied to organic crystal
systems. In Cineca I work as HPC User Support for ENI,
installing software and assisting users on ENI clusters.
Laura Bellentani
I obtained my PhD in Physics and Nanosciences
at the University of Modena and Reggio Emilia. My
postgraduate and postdoc research involved the
theoretical study and numerical simulation of quantum
logic gates in condensed-matter systems. I am working
in the HPC Department of Cineca on the porting and
optimization of codes, mainly focused on Material
Science, and on the support of profiling tools.
Davide Giosué Lippolis
MSc in Theoretical Physics at the University of Bologna,
Alma Mater Studiorum with a thesis in stochastic
modeling and statistical analysis of cytokines metabolic
cycles. He researched in medical image analysis at
Ospedale S.M.N of Reggio Emilia. In Cineca he’s a
desktop and web developer for scientific applications.
Furthermore, he manages DevOps operations on
Kubernetes.
Fabio Di Sante
I got my PhD in “Earth Science and Fluid Mechanics” at
the University of Trieste. I have nine years of experience
in working with hydrological and climatological
models. At Cineca I work in the Data Management
group as support to weather forecasting chains and as
high-level user support to weather and climatological
applications, models and tools.
Giorgia Frumenzio
I got my master in “Molecular Biology” at the University
of Parma. In my research, I studied the dynamics of
several proteins from humans and SARS-COV2 by using
computational techniques like Molecular Dynamics. In
Cineca, I work as HPC specialist providing support for
the optimation of scientific applications.
Lucía Rodríguez Muñoz
I earned a Bachelor’s Degree in Physics, and a Master’s
Degree and a PhD in Astrophysics at the Universidad
Complutense de Madrid (Spain). Then, I worked for 5
years as a researcher at the Università degli Studi di
Padova. The purpose of my research was to improve
our knowledge on galaxy evolution. To this aim, I
analyzed multiwavelength data obtained with some of
the most powerful telescopes in the world to study the
properties of distant galaxies, and subsequently, derive
their most likely evolutionary paths. Currently, I work
as an HPC Data Engineer in the HPC Data Management
group at Cineca.
Elisa Alberti
Elisa Alberti graduated in Political Sciences from the
University of Bologna (Italy) in 2004 and achieved a
specialisation in European Funds from the University
of Padova (Italy) in 2010.
She is a Project Manager with a decade experience in
European Funded projects, in proposal writing and
in financial and administrative mangement, under
different EU funded programmes, such as FP7, H2020,
LIFE, HEU.
She is currently member of PMO (Project Management
Office) of the Cineca’s HPC Department.
Francesco Finelli
I got a master’s degree in Physics at the University of
Rome “La Sapienza” (theoretical Physics, with a focus
on classical dynamics and plasmas) and a Ph.D. in
Physics at the University of Pisa. My research focused
on numerical simulation of magnetic reconnection in
space plasma, both to evaluate models and to generate
data to train and test algorithms for automatic detection
of reconnection in turbulence. In Cineca I work in HPC
User Support.
Maria Montagna
I got a master degree in Physics at La Sapienza
University of Rome, and a PhD in Engineering and
Physical-Mathematical Modelling at the University of
L’Aquila. The main interests of my research activity
were focused on material science and biophysics
(Organic/Inorganic Interactions in Bioinspired
Systems, Catalytic Complexes, Hydration Processes)
using classical and ab-initio molecular dynamics
simulations and theoretical chemistry calculations.
In Cineca I am part of HPC User Support working on
setting and managing of HPC production environment.
Federico Tesser
I got my PhD at the Research Center INRIA Bordeaux -
SouthWest,resolvinginparallelthePoissonequationon
some hierarchical and superimposed grids, in Python.
After that, I came back to Italy, where my occupation
for almost 3 years has been to be a technologist at
the Politecnico di Torino, working on inter/intra node
optimization and performance engineering.

27
Events
The European Resear-
chers’ Night of Society
project
Simona Caraceni
Cineca
In 2021 30 pre-events (24 onsite or in mixed
mode and 6 fully online) were organized in the
month preceding the ERN. The relatively good
Covid situation in Italy allowed us to organise
the majority of pre-events onsite, although
the participants were limited in number by
health safety rules. Most pre-events and the
main ERN event were included in the summer
event program promoted by the Municipality
of Bologna “Bologna Estate”. This gave good
visibility to the project and contributed
positively to the awareness campaign.
Researchers from the Society consortium made
themselves available to give face-to-face or
online lectures in the schools. The rich program
of available lessons (a total of 74 for all school
levels) was advertised directly to teachers and
also through the channels of the Regional
School Office of the Ministry of Education.
Schools in Italy start mid-September and this
leaves too little time to plan lessons before ERN.
Furthermore, this year, schools had to face quite
challenging problems for returning students
to attendance after last year’s remote teaching
period. For these reasons most lessons have
been scheduled to take place in the period after
ERN, preferably by the end of October, but even
beyond.
SOCIETY Rinascimento organized activities
and events for the European Researchers’ Night
in 2021 in seven towns in the Emilia Romagna
Region, Italy: Bologna, Ravenna, Forlì, Cesena,
Rimini, Predappio and, for the first time, Ferrara.
Given the difficulty to predict the evolution
of the pandemic and of the corresponding
restrictions, the consortium decided from the
very beginning to opt for a dual approach,
organizing activities and events both online and
in the streets.
The SOCIETY Rinascimento project organised
a mixed online/onsite event, featuring 37.5% of
activities in presence and 62.5% online. A total
number of 384 researchers were involved, 10 of
whom benefited from MSCA funds (Table 1),
while 91 researchers were supported by other
EU schemes.
For Bologna, the Night in person was organized
downtown, along via Zamboni, at the heart
of the academic district. The road was closed
with identifiable entry points at both ends,
and a single exit, to allow an ordered visting
experience. Access was only granted to holders
of the EU digital Covid certificate released upon
vaccination against COVID-19.
Stands were organized underneath the portico’s
arcades that recently became part of the
UNESCO heritage. Each stand featured a desk
for materials on display and a monitor. The
total number of stands (33) is smaller than in
previous editions. This was necessary to ensure
sufficient space around each activity and also
because some of the activities were proposed
online.
Online activities were available on the virtual
platform of the SOCIETY Rinascimento
project, accessible through the website http://
nottedeiricercatori-society.eu/. Contents of the
virtual stands will remain available to visitors for
one year, until September 2022.
The platform hosted the live streaming of the
inauguration ceremony, thanks to the support
of our stakeholder Lepida TV. The possibility of
broadcasting further live events held in presence
was discussed among the partners but eventually
discarded, to focus the visitors’ attention to the
many virtual stands held by the researchers.
We estimated that approximately 4700
people attended all events. Considering the
pandemic limitations, we registered a good
onsite attendance in Bologna (approximately
2000 participants), Rimini (354), Ferrara (330),
Ravenna (231), Forlì (144). Online activities
were followed through the sharevent platform
(723 registered users) or through social media
channels such as youtube and Fb (965). The
web page where the scientific magazine Sapere
was made available for download had more than
800 accesses, both during the Night (183) and
afterwards.

28
Figures 15, 16: European Researchers’ Night in Bologna
Figures 17, 18, 19: European Researchers’ Night in Rome

29
Events
The European Resear-
chers’ Night of Net project
Neva Besker
Cineca
The NET- scieNcE Together project organized
its second edition of the European Researchers’
Night, an initiative promoted by the European
Commission since 2005 that involves thou-
sands of researchers and research institutions
every year in all European countries.
The topic was the challenges given by climate
change. The goal was to create opportuni-
ties for researchers and citizens to meet and
spread scientific culture and knowledge of the
research professions in an informal and stimu-
lating context. NET is a partnership of all the
major local (Lazio) Research Institutes and
Universities, such as CNR (coordinator of the
project), Cineca, Ispra, INAF, INGV, ENEA,
University of Rome Sapienza, University Tor
Vergata, University of Tuscia and Telematic
University UniNettuno. Cineca in this 2021
edition coordinated the Impact Assessment.
The impact assessment campaign was designed
to contribute towards achieving the objectives
of assessing the impact of the NIGHT event,
using qualitative and quantitative parameters,
and of identifying what kinds of activities are
most successful in attaining the European Rese-
archers’ Night main aims.
The project had an ambitious program of a
total of 70 pre-events with the participation of
more than 2.800 people to promote the Rese-
archers Night.
Apart from Rome, where most of the activities
took place, the pre-events were in many cities
involved in NET project such as Livorno, Vul-
cano, La Spezia, Viterbo, Cerveteri, Tarquinia,
Chioggia, Ozzano, Brasimone.
The pre-events were conducted in the most
diverse contexts to bring researchers into closer
contact with citizens, outside the traditional
context of scientific research, promoting the
interaction.
NET scientific treks achieved great success
with 36 events organized in many locations
throughout the national territory: naturalistic
treks or in urban parks, urban walks visiting
churches and finding links between art and
astronomy, trekking on the volcanoes of the
Aeolian islands, were an opportunity to discuss
scientific disciplines such as botany, biology,
geology, astronomy, archaeology and more.
NET 2021 promoted the 7 “scientific aperitif”
format bringing researchers in very informal
contexts in the late afternoon, in close contact
with an audience of 40-50 people. The loca-
tions were associations – Casetta Rossa and
Fusolab 2.0 - in two districts of Rome, one in
the centre and the other in a suburban area,
both frequented by young people who engaged
in social activities.
Another important pre-event was “Talenti per
la scienza” at the Talenti Park Arena, lasting 2
days, in which hands-on activities for children,
demonstration stands, scientific talks and show
conferences were organized, with the presence
of researchers of almost all NET partners.
NET 2021’s most prestigious pre-event fea-

30
tured the 2019 Nobel Prize in Physics Michel
Mayor. Professor Mayor was a project guest for
three days in Rome at Sapienza University of
Rome.
The presence of Prof. Mayor was an opportu-
nity to explore the themes of space research
and extraterrestrial life, considering that Mayor
discovered the first exoplanet orbiting a so-
lar-type star in our galaxy, the Milky Way.
The activities carried out for the NET 2021
Night included a program dedicated to studen-
ts of Italian primary and high schools mainly
from the Lazio region with the participation
of various institutes from other Italian regions.
The activities for schools were launched with
an initiative called “Pillole di scienza-Pills of
Science (Did you know that ...)”, short videos
dedicated to a specific scientific theme. The
videos were produced by each Partner on some
themes which were then discussed in subse-
quent lessons. Many activities were converted
into online activities
The in person and online activities for the
schools took place mainly in the period from
20 September to 25 September 2021, during
the week of the Night with an extension until
November 2021. The activities dedicated to
schools can be summarized in 108 events. 53
events over 108 were carried out in presence
(32 at partner sites, research centres and event
venues of the Researchers’ Night and 21 direct-
ly at the School sites). The remaining 55 events
were online events (39 live events with the
researcher connected to the classes and 16 on
demand, always available on the internet).
The NET Consortium organized the final event
principally in the centre of Rome (Testaccio)
inside the “The City of the Other Economy -
CAE” which is one of the first spaces in Europe
entirely dedicated to the promotion of the
other economy. CAE is a very big outdoor pla-
ce where thousands of people could be hosted
even according to the strict anti-covid rules
that were forced in Italy in September 2021.
The final event was organised for two nights 24
and 25 September (Friday and Saturday) from
18:00 to 24:00. In addition to this big event,
those partners who could host the general
public at their sites arranged also other events.
Moreover, some online activities have been
done, such as updated virtual tours for almost
all institutions of the consortium to open vir-
tually their scientific laboratories that were not
accessible during the pandemia. It was a way to
visit the environment of laboratories in which
the researchers carry out their research daily
and to permit them to discover, even from
home, the tools and technologies used.
Researchers involved in the final event inside
CAE welcomed about 4000 visitors with a
wide offer of activities and about 30 stands.
Many activities were designed for the families
and general audience offering hands-on activi-
ties, games, science demonstrations, informal
conversations with researchers and science
entertainment. The participation in the other
cities was about 1500 attendees. More than half
of the visitors attended for the first time events
related to the European researchers Night. The
events had a great impact on the participants,
who declared, overall (more than 97% in all
questionnaires ), to be interested in attending
similar events. These numbers are significant to
describe the satisfaction and the effectiveness
of the NET events.

32
Figure 20: the winners of the award
Events
Summer of HPC 2021
Massimiliano Guarrasi
Cineca
Also in 2021, Cineca participated to PRACE Summer of HPC program.
Summer of HPC (SoHPC) is an initiative funded by the European Union through the PRACE
project, which aims to allow University students from all over Europe to have a study and work
experience abroad at major European computer centres. The selected students work under the
supervision of a mentor (generally a researcher) and a company tutor (coming from the computer
centre that will host them), on projects related to High Performace Computing (HPC) and scientific
visualization using calculation resources provided by PRACE.
PRACE Summer of HPC aims to increase the interest of new generations in HPC and related
technologies by disseminating the experiences of program participants via social media.
Since the pandemic affect also the possibility to go abroad on 2021, the internship had to take
place remotely. Cineca hosted four students working on two separate projects. Particularly David
Mulero Perez and Sepideh Shamsizadeh worked under the supervision of Prof. Andrea Bartolini
of the University of Bologna on a project titled “Visualisation and Anomaly Detection of A
Supercomputer”.
The other project was coordinated by Prof. Silvia Ceramicola of National Institute of Oceanography
and Experimental Geophysics (OGS). The two students that work on this project, Mario Gaimann
and Raska Soemantoro won also the 2021 SoHPC Best Performance Award, with the project
“Automated Classification for Mapping submarine structures by Artificial Intelligence strategies”.
The award ceremony took place during the EuroHPC Summit Week 2022 / PRACEdays22 on
March 24th, 2022 in Paris. The two students worked in collaboration with OGS, NOC and Cineca,
as well as the DPC-MaGIC Project.
The reasons for the award were: “Mario and Raska worked on a particularly challenging project
and on their own developed an automated tool to recognize seabed structures. They worked very
hard on this project, researching the various AI methods available to them and deciding on which
methods were best to use. Furthermore, through their own initiative they researched, identified
and learned various technologies and tools to use for their project. The outcomes of their work
far exceeded the initial expectations of the project. Indeed, they will aim to publish their results
through funding via the HPC-Europa3 scheme.”

33
W e b o f S c i e n c e
W e b o f S c i e n c e D a t a b a s e : C o r e C o l l e c t i o n
S e a r c h : ( F T = C i n e c a o r F O = C i n e c a )
D a t e :
S o u r c e ;
a n d ( P Y = 2 0 2 1 - 2 0 2 2 )
0 3 . 1 1 . 2 0 2 2

34
Figures 21, 22: number of publications
and citations from year 2008 to 2022
(2022 only January-October)
Figure 23: research areas of the publications

35
Figure 24: main affiliations of the authors
Figure 25: journals in which Cineca users published most frequently
Figure 26: nationalities of the organizations of authors

37
Cineca is involved
in a wide scientific
field thanks to
EU projects and
collaborations.
From artificial
intelligence to
urgent computing,
bioinformatics and
digital humanities,
HPC is used to support
them all.

39
ORCHESTRA
Connecting European Cohorts to Increase Common and Effective Response to SARS-CoV-2
Pandemic
Gabriella Scipione, Chiara Della Casa, Elisa Rossi
Cineca
Figure 27: ORCHESTRA partners map
ORCHESTRA is a three-year international
research project aimed at tackling the
coronavirus pandemic, led by University
of Verona and involving 26 partners (which
extends to 37 partners when considering
the wider network) from 15 countries.
The project, started in December 2020 with
a budget close to 20 million euro, is funded
by the European Union’s Horizon 2020
research and innovation programme.
The ORCHESTRA project aims to respond to
this need through the creation of a new pan-
European cohort built on existing and new
large-scale population cohorts in European
and non-European countries.
Objectives
•todevelopevidence-basedrecommendations
for effective prevention, protection and
optimized treatment of COVID-19 patients
(including long-term consequences) with
a special focus on ‘at risk’ population,
including healthcare workers and fragile
individuals
• to assess impact of environmental factors,
socio-economic determinants, lifestyle and
confinement measures on the spread of
COVID-19
• to provide knowledge on the efficacy of
vaccines against SARS-CoV-2
• to provide a model for responsiveness for
future pandemic outbreaks.
In this context, the integration and analysis
of the very heterogeneous characteristics of
SARS-CoV-2 health data coming from many
different sources such as electronic health
records, retrospective and prospective
patient registries, and related ‘omics’ data,
are key drivers for the evolution from
evidence-based medicine towards precision
medicine and precision public health.
Cineca in the project is the coordinator of
the WP7 “Data Management” work package,
in charge of the development of the platform
thatwillcollectdatafromthevariouscohorts
involved. Together with other leading HPC
European Centers (HLRS - Germany and
CINES - France), Cineca will be responsible
forthedesignanddeploymentof afederated
architecture based on three layers: National
Data Providers, National Hubs (NH) and the
centralized Orchestra Portal.

40
ORCHESTRA has received
funding from the Europe-
an Union’s Horizon 2020
research and innovation
programme under grant
agreement No 101016167.
Figure 28: ORCHESTRA objectives
Starting from the definition of the general
architecture, CINECA deployed a first
solution to manage sensitive personal
data in the Italian National Hub (NH) in
compliance with the GDPR and the Italian
regulations.
Data are ingested from the national data
providers, mainly from Universities and
Hospitals located in Italy, using a dedicated
instance of REDCap provided in the NH
infrastructure.
The process includes managing and storing
pseudonymized data in the National Hub
(NH) based on a cloud infrastructure with
typical repository functions. Before opening
the NH to real world data collection, some
actions have been carried out at Cineca:
- signature of the “appointment of data
processor”, a data sharing agreement
between each data controller (National Data
Provider) and data processor (CINECA),
- adoption of all technical and organisational
measures to minimize the risk,
- implementation of a Vulnerability
Assessment and Penetration Testing (VAPT)
- execution of a detailed internal audit, with
internal DPO officers,
- Preparation of a document to support Data
Protection Impact Assessment (DPIA).
Additionally, for all the use cases in which
standard data sharing procedures are not
applicable by definition or due to any kind of
legal restrictions (e.g. informed consent to
share data for research purposes not given
for retrospective cohorts), alternatives such
as Federated Learning are explored.
Considering the European context, in
Europe there is an increasing demand for
regulated data usage in HPC data centres
with respect to GDPR regulations. The
HPC cloud infrastructure released for the
Orchestra project can act as a model to
manage future public health threats.

41
ARNO
Elisa Rossi, Salvatore Cataudella
Cineca
Starting from 1987, ARNO Observatory routinely collects and integrates NHS administrative
data for each single patient, to support governance activities in the healthcare (expenditure
control and appropriateness of drug prescription) of LHUs and to study Patient Care
Pathways.
ARNO Observatory allows to highlight cohorts of population with chronic disease that need
to be followed, evaluated, and studied. Diabetes is expected to become one of the world’s
most common diseases within the next twenty-five years with high impact on National
Health System resources, it is a condition with high prevalence of comorbidities, and which
involves high quantity of resources. The aim of ARNO Diabetes Observatory, in collaboration
with SID (Società Italiana di Diabetologia), is to assess the impact of resources of patients
with diabetes, compared to people without diabetes, in order to analyze prevalence and
incidence of this disease and evaluate the burden of disease in terms of impact of new drugs,
comorbidity, complications and costs. ARNO Diabetes has become one of the reference
sources for the scientific community and periodically Cineca publishes reports to describe
the patient profile and use of heathcare resources (Figure 1).
In 2021, new papers related to initial treatment and to incidence of diabetes have been
published, respectively to:
1) investigate diabetes treatment initiation and continuation in the next sixth months in
newly diagnosed Italian subjects1
2) assess incidence of diabetes in Italy from several million residents, highlighting
differences in rates in men and women across decades of age and Incident rates of insulin
or noninsulin treated subjects2
.
Figures 29-33 reports on ARNO Diabetes
2007 2011 2015 2017 2019
1 Bonora E, Cataudella S, Marchesini G, Miccoli R, Vaccaro O, Fadini GP, Martini N, Rossi E; mandate of the
Italian Diabetes Society. Initial treatment of diabetes in Italy. A nationwide population-based study from of
the ARNO Diabetes Observatory. Nutr Metab Cardiovasc Dis. 2021 Aug 26;31(9):2661-2668. doi: 10.1016/j.
numecd.2021.06.006. Epub 2021 Jun 22.PMID: 34218990
2 Bonora E, Cataudella S, Marchesini G, Miccoli R, Vaccaro O, Fadini GP, Martini N, Rossi E; under the mandate of
the Italian Diabetes Society. Incidence of diabetes mellitus in Italy in year 2018. A nationwide population-based
study of the ARNO Diabetes Observatory. Nutr Metab Cardiovasc Dis. 2021 Jul 22;31(8):2338-2344. doi: 10.1016/j.
numecd.2021.04.017. Epub 2021 May 3.PMID: 34074587

42
EARTH AND
ENVIRONMENTAL
SCIENCES

43
SEBASTIEN
Smarter livEstock Breeding through Advanced Services Tailoring Innovative and mul-
ti-sourcE data to users’ Needs
Chiara Dellacasa, Cinzia Caroli
Cineca
Paola Nassisi, Antonio Aloisio
CMCC
Climate change has an increasingly
important impact on livestock farms,
influencing production, health and fertility
of animals. At the same time, farms are
sources of greenhouse gases, at the root of
global warming. It is therefore necessary
to develop strategies that help the entire
livestock sector to make the best decisions
to adapt to these changes and mitigate their
effects.
SEBASTIEN is a new project co-funded
by the Connecting Europe Facility of the
European Union, and it aims to deliver a
DecisionSupportSystem(DSS)intheformof
a package of services hosted in an open and
interoperable web Portal. The main goal of
the project is to increase the efficiency and
sustainability of the Italian livestock system
and support breeders and operators in the
animal production chain with tools that
can help them make targeted and informed
choices, but also avoid economic losses and
provide technical support to companies.
SEBASTIEN started at the beginning of 2022
and is coordinated by CMCC Foundation
(https://www.cmcc.it/); it relies on Cineca
and CMCC infrastructure for monitoring
data expolitation and on climate simulations
already produced by the Highlander project
coordinated by Cineca. The applications
developed by SEBASTIEN will use large
amounts of public data, of different nature,
harmonized and analysed using the most
modern analysis techniques such as
artificial intelligence methods.
TheSEBASTIENprojectwilldevelopsentinel
indicators by integrating environmental,
meteorological, reproductive and
production data:
- In situ sensors that allow measuring
parameters related to animal body
conditions.
-Historicalandfutureclimatesimulationsat
unprecedentedspatialdetailsoverItaly,tobe
combined into indicators of environmental
comfort/wellbeing conditions.
- Satellite images integrated for multi-
dimensional analyses of feed availability.
Four free applications will be developed
which will allow:
1. estimating the adaptation of different
livestock species and breeds to changing
environmental conditions
2. providing a forecast system that allows
farmers to take actions to fight dangerous
environmental conditions
3. guiding farmers in the choice of the best
pastures
4.monitoringthepresenceanddevelopment
of parasites and pathogens.
The indicators will be tailored to different
categories of end users, to allow breeders,
companies, government bodies and other
operators in the agri-food-livestock sector
to be able to face their daily needs and work
problems quickly and efficiently.
Co-financed by the
Connecting Europe
Facility of the Europe-
an Union
Grant agreement
n° INEA/CEF/ICT/
A2018/1815462

44
Digital technology enablement for Desti-
nation Earth
Technology Partnership Lead for Destination Earth
Compute Department
Thomas Geenen
ECMWF
ECMWF (the European Centre for Medium-
Range Weather Forecasts) is one of three
entrusted entities tasked with delivering
Destination Earth (DestinE). DestinE is
an ambitious initiative of the European
Commission to develop a highly accurate
digital twin, or replica, of Earth. This will
allow users of all levels to better explore
natural and human activity and to develop
and test a range of scenarios and potential
mitigation strategies. For example, it
will better enable policy and decision
makers to anticipate and mitigate the
effects of extreme weather events and
climate change;potentially saving lives and
alleviating economic consequences.
Bringing together scientific and industrial
excellence from across Europe, DestinE will
contribute to revolutionising the European
capability to monitor and predict our
changing planet, complementing existing
national and European efforts such as those
provided by the national meteorological
services and the Copernicus Services.
Destination Earth will also support the
EuropeanCommission’sDigitalStrategyand
the Green Deal priority actions on climate
change, biodiversity and deforestation.
High-performance computing (HPC)
and big data handling have been a key
enabler for weather and climate prediction
for decades. The recent evolution of
computing architectures has stimulated
a wide range of programmes that aim to
prepare the operational weather prediction
infrastructures for future technologies.
These programmes increasingly realize
that it will take more than progress in HPC
to fulfil future extremes prediction and in
particular climate adaptation targets as
addressed by DestinE. This is where the
need to exploit the opportunities offered
by the entire range of digital technologies
(Figure 1) comes into play, and where new
ways of co-developing Earth system and
computational science, and co-developing
DestinE’s digital ecosystem need to be
found.
Cineca has been selected to support
ECMWF on this journey, leveraging decades
of experience in HPC and code optimization
and as one of the leading HPC centres in
Europe, deploying the first generation
pre-exascale systems in Europe. Code
optimization and cloud experts from Cineca
will work closely with ECMWF developers
and join the workstreams tasked to deliver
technical enablers that builds the evolving
architectural runway supporting the digital
twin deployments. Since the ultimate
ambition of DestinE is to run digital twins
across all EuroHPC systems, a large focus
will be in portability of applications and
services to all (pre) exascale systems. We
will also leverage the extended experience
of Cineca in community management and
training to onboard the DestinE community
and associated consortia on the EuroHPC
ecosystem.

45
Figure 34: DestinEs extreme-scale capabilities critically depend on the seamless
interplay with many components of the digital continuum. Cineca is well posi-
tioned to both work on the capabilities side, scaling simulations and models to
exascale system size, and on several components of the digital continuum.

46
BIG DATA
AND
ARTIFICIAL
INTELLIGENCE

47
Ocean, Supercomputers and AI in action!
Roberta Turra, Giorgio Pedrazzi, Bhaskar Agarwal
Cineca
Marcello Maranesi, Matteo Picchiani
Gmatics
AI models engineered on Cineca HPC
infrastructure are a core element of the
services that Gmatics developed as part
of the DAMAS – Data-driven Model for the
Analysis of Sea-state experiment funded
by the EUHubs4Data project (https://
euhubs4data.eu/experiments/damas/).
The experiment successfully reduced the
execution time and increased the resolution
and the accuracy of sea circulation models
and height of waves prediction, providing as
a result:
• AI models that replicate the ENEA MITO
and WAVE physical forecast models and
produce forecasts over the next 5 days with
hourly temporal resolution,
• The nowcast chains for wave and
circulation for the next 12 hours,
• Hindcast pipelines for the historical
analysis of wave and circulation sea
parameters and
• A graphical user environment
122.586 core hours have been consumed
and 44 TB of data have been processed in
order to reduce the AI model runtime for the
whole Mediterranean Sea to 15 minutes (the
experiment final target was 30 minutes),
increase the geospatial resolution to 1x1
Km, achieve a point accuracy with a Root
Mean Square Error between DAMAS and
ENEA results below 5 % for each variable
(total peak and mean period, total mean
wave direction, sea surface wave significant
height, sea surface height, surface salinity,
surface temperature, surface wind
u-component and v-component speed) and
increase the areal accuracy above 90% for
each variable.
These results were achieved during the 9
months duration of the experiment by first
matching the performance of the ENEA
model and then improving the geospatial
and temporal resolution, which required
the fine tuning of the AI model using satellite
data, ensembling techniques and transfer
learning techniques.
DAMAS was funded by the EUHubs4Data
project (https://euhubs4data.eu/ ) that
promotes the collaboration between Data
Driven Innovation Hubs in EU through the
Figure 35: DAMAS dashboard showing forecasts on two variables
(wave direction and wave height) at a specific location of the
Mediterrean sea

48
definition of a common catalogue of datasets, data services and training services. Towards
this goal, SMEs are invited to propose experiments that make use of the EUHubs4Data
federated catalogue for the development of innovative products or advanced services and
are granted access to funding through the implementation of open calls. Cineca is one of
the participating Innovation Hubs providing services to the SMEs and oversees the training
programme and the project Data Management Plan.
A Federation of Data Driven Innovation Hubs is being established in order to be operational
during the project and beyond as the European reference for data driven innovation
and experimentation, fostering collaboration between data driven initiatives in Europe,
federating solutions in a global common catalogue of data services, and sharing data in a
cross-border and cross-sector way.
The EUHUBS4DATA project
has received funding from
the European Union’s
Horizon 2020 research and
innovation programme
under grant agreement No
951771

49
AI Day-ahead hail prediction to prevent
losses in agriculture
Bhaskar Agarwal, Michele Bottazzi, Giorgio Pedrazzi, Gian Franco Marras, Roberta Turra
Cineca
Every year, almost a third of all food
produced is lost or discarded, resulting in
a 940 billion economic loss. This comes at
a time when it is estimated that 50 percent
more food will be required over the next
20-30 years. The increased prevalence
of extreme weather events due to climate
change has heightened the need to build
support decision systems that can assist
farmers in mitigating agricultural losses.
Precision Agriculture is a technology-
enabled, data-driven approach to farming
management that watches, measures, and
analyses the demands of fields and crops to
help optimize agricultural production while
reducingwasteandexpenses.TheEUCybele
project (www.cybele-project.eu/) aims at
generating innovation and creating value in
the domains of agri-food by implementing
such Precision Agriculture demonstrators.
Our task within the project was to develop
a hail prediction model. Hail is a rare
phenomenon, very difficult to predict. We
treatedtheproblemasabinaryclassification
to predict whether it will hail or not at a
given hour and location, a day in advance.
The focus was on the Valencia region, and
an enclosing area of around 200,000 square
kilometers in Spain was selected for training
the model. We used the weather forecast
and the derived climate instability indices
as input, amounting to a total of 6 TB. The
location and timestamp of 100 hail events
in Spain was available as truth data for the
periodof2015-2019.Owingtotheimbalance
in the problem, we decided to spread a hail
event both temporally (an hour in the past
and the future) and spatially (a radius of
up to 10 kms). This led to a more physical
and complete target data set. We employed
3 different gradient boosting algorithms
to create an ensemble model and worked
with two KPIs. While we optimised the
model based on the F1-score, we assessed
the Recall and `Corrected Recall’ of the test
set. The Corrected Recall, a completely new
KPI, was designed for the project based
on synthetic target data creation. After
assessing the needs of the stakeholders, it
was clear that the hail events most critical
to them were the summer months. Thus, we
trained and optimised the model in these
months from 2015-2017 and tested on the
same months in 2018-2019.
Our model returned a lower than 20 % false
positive rate, and a 50-100% Corrected
Recall of hail events. The final model was
theningestedincontaineriseddemonstrator
designed for direct use by the farmers.
The EUHUBS4DATA project
has received funding from
the European Union’s
under grant agreement No
951771

50
Figure 36: model diagram CYBELE project

52
MaX: HPC and HTC meet for Material
Sciences
Materials at Exascale
Mariella Ippolito, Fabio Affinito
Cineca
This year will see the conclusion of the
second MaX project funded under the
Horizon2020 programmme of the European
Commission. During the last years, thanks
to MaX, Cineca took part in an activity aimed
at make the codes for electronic structure
calculations ready for the supercomputers
of next generation, including the incoming
Leonardo machine. In this framework,
Cineca was particularly involved in two
different actions. The first one was about
the porting and the optimization of the
Quantum ESPRESSO code on the GPUs
accelerators. In collaboration with the
SISSA team (where the code was born),
we actively participated to transform
the code the code so that it could run on
accelerated systems, like Marconi100. This
allows the material scientists to exploit at
the best the most modern computational
architectures, getting advantage from the
computing power offered by the NVIDIA
accelerators. The second action, which has
seen CINECA as leader, is the coordination
of the user support in the MaX project. This
role permitted to help the community of
users to fix problems related to execution of
complex applications on HPC systems. The
role of the user support for material science
applications will be furtherly enhanced in
the next years, also beyond the MaX project,
with the implementation of the application
support in the High-Level Support Team
of the Leonardo system. Looking ahead
at the next future, the availability of large-
scale infrastructures, such as Leonardo,
will open new possibilities for many
different scientific domains, including
material sciences. As a matter of fact, we
expect that besides a single application
that is able to run on a very large number
of computing nodes, we will see a different
model of exploitation, where a multiplicity
of application running on few nodes will
interact together for the resolution of a
physical problem. In other words, we expect
that HPC, High-Performance Computing
(a few applications with large scale-out),
will be tightly working with HTC, High-
Throughput computing (a large number
of applications with small scale-out). The
management of complex workflows is really
challenging from a technical point of view,
because it requires a high robustness of
the computing infrastructure including an
efficient management scheduling system.
The experience and the expertise obtained
in these years with MaX will permit us to
getting ready to prepare the new challenges
posed by the advent of Leonardo and to
provide the material scientists with new
complex tools to discover the materials of
our future.
MAX has received funding
from the European Union’s
under grant agreement No.
824143.

55
Cineca Quantum computing perspectives
Daniele Ottaviaani
Cineca
Without a doubt, the concrete realization
of the first quantum computers represents
one of the most important scientific events
of this new millennium. Theorized at the
beginning of the 1980s, for over 40 years
scientists all over the world have been
studying how to make devices capable of
exploiting the laws of quantum mechanics
to increase our computing power.
After 20 years of studies, it was finally
possible to create some small prototypes
of a functioning quantum computer in
the laboratory: for the real breakthrough,
however, it is necessary to wait another
twenty years, to arrive at the present day.
Today quantum computers are gradually
approaching the world of supercomputing.
Although far from the science fiction models
of universal quantum computers theorized
in the 1980s, the computers we have today
are beginning to be suitable to be used
effectively as accelerators for computing,
as well as being irreplaceable scientific
research tools.
A quantum computer is a device which
is completely different to a traditional
computer.Perhapstheonlythingincommon
is the fact that, given an input, it manages to
complete calculations in order to produce an
output. But the ways in which it manages to
manipulate the input data in order to return
a result are completely different to those
used by a classic computer. The quantum
computer, in fact, bases its computation
on a basic logical unit that in some sense
can be considered an evolution of the bit,
called qubit. A qubit (short for quantum
bit) is basically a bit capable of exploiting
the laws of quantum mechanics to assume
superposition states between the two
classical states 0 and 1, effectively offering
the theorists of quantum algorithms the
possibility of exploiting effects impossible
to play with a classic computer.
Thepowerofquantumcomputersisprecisely
this: given their particular nature very close
to the laws of quantum mechanics, they
offer developers the possibility of writing
very advanced algorithms, with some “extra
gears” that allow them, in some cases, to be
enormously better performing than their
classic counterparts. I say “in some cases”
because it is absolutely not certain that a
quantum algorithm is necessarily better
than a classical algorithm: it depends on
the problem you want to try to solve. This
statement also gives me the opportunity
to underline a very specific concept: a
quantum computer is not a more powerful
classical computer, but rather a different
device. A quantum computer will never
replace the classical computer: although
there are several algorithms capable of
guaranteeing even exponential speed-ups
compared to classical algorithms, there are
also many other fields of application where
the classical computer is more performing
than a quantum computer.
Quantum computing is the science that
explains how to program a quantum
computer; given the deep differences
between classical and quantum computers,
quantum computing also has profound
differences with classical computing.
Learning to program and to use the new
quantum devices in the best possible way
and as soon as possible is strategically very
important, especially for a supercomputing
center.
It is now clear that quantum computing is a
branch of computer science closely linked
to high performance computing, especially
if we look closely at this historical period. In
fact, in recent years we are about to reach
whatscientistscalltheeraofNISQcomputers
(Noisy Intermediate-Scale Quantum
computers). A NISQ computer, exactly as
its acronym says, is a quantum computer
of intermediate scale (whose chipset varies
between a hundred and a thousand qubits)
and noisy (i.e. the quality of the qubits is
not yet sufficient to run the most famous
algorithms. and more demanding in terms
of hardware). Its importance lies in the fact
that, according to a large part of the world
scientific community, it will be one of the
first devices with which it will be possible to
reach the first forms of quantum advantage

56
(i.e. the moment in which a quantum computer will be able to perform more faster than
classic supercomputers).
More specifically: a NISQ computer will be able to achieve quantum advantage thanks to its
use in synergy with current supercomputers. Put simply, we think of using NISQ computers
as accelerators for computing, in the same way as GPUs are used today.
We at Cineca have been monitoring the evolution of quantum computers since 2018. Over
the years we have studied and tested quantum technologies on the market, as well as
having studied quantum computing also by exploiting our internal computing resources to
create quantum computer emulators. In 2021 we founded the Quantum Computing Lab, of
which I am the coordinator, and since then we have worked hard to try to offer our Italian
universities and research centers all the help we can to understand and exploit this new
technology. At the moment we can divide our commitment on three fronts:
Collaboration with European and Italian supercomputing centers and universities for the
development of quantum computing: the European scientific community is very active from
the point of view of software development for quantum computing. We at Cineca collaborate
with other European players on various projects financed by the European community and
by Italian internal resources, such as the PNRR. In particular, it is worth mentioning the
HPCQS consortium, of which we are an active part, which was born in response to a call from
the EuroHPC JU for the construction of a quantum computer network capable of supporting
European supercomputers. From the italian side it is worth mentioning our participation
in at least two partnerships concerning the development of quantum software and financed
by the PNRR. Not to mention all the other projects that we have submitted and of which we
are waiting to know the outcome.
In addition to the partecipation in the European HPCQS project, Cineca is also one of the
founders of the EuroQCS (European Quantum Computers and Simulators) initiative. The
EuroQCS initiative was born in February 2022, with the drafting of a whitepaper with
the same name in collaboration with the major European supercomputing centers. The
whitepaper focuses the reader’s attention on the dual problem of integration between
HPC and QC and on the need to be able to access the largest number of QCs available on
the market today. In fact, unlike traditional classical computers, the choice of technology
with which to build the qubits of quantum computers greatly influences their performance
(and we don’t know yet which is te best one!). The EuroQCS initiative became even more
concrete in October 2022, when the ranking for the acquisition of a QC through EuroHPC
JU was made known. Italy, represented by Cineca, participated and won under the name of
EuroQCS-Italy and together with other hosting entities such as France, Spain and Poland as
well as partners such as Germany, Slovenia and Ireland, officially gave birth to the EuroQCS
consortium, preparing to give the possibility for all European researchers to be able to use
in a simple and effective way all the main quantum technologies currently on the market.
Teaching and Outreaching: in addition to actively working on the development of quantum
computing software, we at Cineca also take care of the creation of schools and dissemination
events to teach future generations of quantum programmers and interested researchers
how to correctly use these new technologies. We started in 2018 with the first edition of
the very popular HPCQC conference (now in its fourth edition); between 2020 and 2021
we created 2 editions of the “Introduction to Quantum Computing” school and one edition
of the “Practical Quantum Computing School”, as well as countless participations in many
dissemination events.
Support to Italian research: to complete the development and teaching activities, we also
offer the concrete possibility to Italian universities and research centers to use via the cloud
various different technologies of quantum computers, offered totally free of charge through
Figure 37: QuantumLab website
https://www.quantumcomputinglab.cineca.it/

58
HUMANITIES AND
SOCIAL SCIENCES,
DIGITAL
ECOSYSTEMS

59
The Great Circle - The immersive experi-
ence of Tübke Monumental
Luisa Mantovani-Löffler, Markus Löffler
Kunstkraftwerk, Leipzig, Germany
“TÜBKE MONUMENTAL” is a digital art
installationpresentedbytheKunstkraftwerk
in Leipzig, Germany. It is based on the world
heritage artwork: THE EARLY BOURGEOIS
REVOLUTION IN GERMANY.
The painting, located in the Panorama
Museum in Bad Frankenhausen, Germany,
is one of the largest panorama oil on canvas
paintings in the world. It spans 14 metres by
123 metres with more than 3000 characters.
It was created between 1976 to 1987 by the
Leipzig painter Werner Tübke (1929-2004),
recognized as one of the most important
East German artist.
The project has tree components. First, a
high-resolution digitalisation of this artwork
has been undertaken, resulting in a stitched
unique 300 Gigapixel image. It is made
explorable on a large touch screen, usable
for research and digital art mediation.
Visitors can easily navigate through the
painting to explore every detail.
Second, to explore the personality of Werner
Tübke, five eye witnesses were invited to
report their encounters with the artist. They
provide a deep insight into his mentality
and mission.
Third, the key part of the project is the 27-
min immersive multimedia installation
named „THE GREAT CIRCLE“ created by
Franz Fischnaller in close cooperation with
VisITLab/Cineca (Antonella Guidazzoli,
Maria Chiara Liguori, Silvano Imboden
e Daniele De Luca) and the american
composer Steve Bryson.
Franz Fischnaller has developed the story
board in 12 scenes, elaborating the painting
together with the VisITLab/Cineca team in a
creative combination of in-between the 2D
gigapixel images and 3D visualisation.
It was a challenge to select 250 figures of
the painting and rearrange them according
to the storyboard. Besides modelling,
texturing, animating, lighting, technical
issues were addressed.
VisITLab/Cineca had a big contribution in
(e.g. the 3D reconstruction of the landscape,
the complex fluid animation of a flooding,
the tower of Babylon which rotates and
breaks into pieces or the devil to operate his
human trap).
The installation provides a unique
emotional and multisensorial introduction
into the painting whose theme is the ever
repeating history of failing utopias and
human incapability to create a social and
just society.
Tübke Monumental was opended to the
public in March. It had an enthusiatic
media and audience response. This has
been praized as setting new standards in
immersive art perception. It will be open
until end of 2022.
,

61
Figure 38: the Great Circle (ph. Luca Migliore)
Figure 39: another perspective of the visual installation (ph. Luca Migliore)
Figure 40: keyvisual of the installation (ph. Luca Migliore)
Figure 41: during the vernissage at Kunstkraftwerk (ph. Luca Migliore)

62
DARE: a digital environment for urban
regeneration
Alessandro Bucci
Municipality of Ravenna - DARE management team
Between 2021 and 2022 Cineca has given great contributions to the UIA European Project
DARE - Digital Environment for Collaborative Alliances to Regenerate Urban Ecosystem
in middle-sized cities. At the heart of the project there is the aim to provide a digital
environment that makes data accessible, understandable, and useful, to enable citizens to
become actors in the ongoing Urban Regeneration Process of the Darsena neighbourhood
in Ravenna, Italy. The digital environment is composed of 3 parts: a Data Management
Platform (DMP), a Content Management System (CMS), and a web interface that functions
as a portal (www.darsenaravenna.it). In this framework, Cineca is responsible for the
development of the DMP, i.e. the very foundation of the digital environment. In fact, the DMP
is the digital space where data from the entire ecosystem of the Darsena is collected and
elaborated, and is designed to receive and retrieve data from various sources, from sensors
to archives. This wide variety of data is here catalogued and made available for subsequent
processing to assess, for example, the level of the Quality of Life in the area. The results
are then elaborated and rendered to be displayed online through the web interface. For
example, the DMP today receives data from sensors installed inside two public apartment
buildings and a school, collecting information on energy performance, indoor and outdoor
air quality and thermo-hygrometric conditions: this provides insights to citizens and policy
makers on which aspects need intervention and what goals to aim for in terms of increased
well being and energy saving.
Through an authentication function, authorised users can progressively define which data
sources are active and select the parameters for information retrieval, as well as view their
history. A system of Rest APIs, developed with the OpenAPI (Swagger) standard, is used to
make the data available to be visualised online.
To grant the durability and the active use of the DMP beyond the project, a pivotal element
is to build the necessary competences within the local Public Administration, which owns
and manages the digital environment. To do so, Cineca on one hand is designing a capacity
building program enabling a selected group of members of the IT department, in particular
of GIS-Geographic Information System Office, to operate the DMP autonomously and, on the
other, is simplifying the use of the DMP, to make the data management increasingly user-
friendly.
Urban Innovative Actions is an
Initiative of the European Union promoting pilot projects
in the field of sustainable urban development.

63
Figure 42: DARE project diagram

64
DARE-IMC platform and the management
of 3D digital objects
Elisa Borghi, Antonella Guidazzoli, Maria Chiara Liguori
Cineca
DARE Darsena Ravenna project - https://www.darsenaravenna.it/ - aims to demonstrate
the effectiveness of an innovative digital-based and citizen-centred governance approach,
to facilitate, support and speed up the implementation and evaluation of the regeneration
process of the Darsena, a neighbourhood in the Italian city of Ravenna.
As part of the project and its effort in involving the citizens, a selection of 3D reconstructions
related to the history of the area was created by FrameLAB, the multimedia and digital
storytelling laboratory of the University of Bologna, rising the need of managing them in the
DARE-IMC platform, a customised version of IMediaCities- https://www.imediacities.eu/ -
for the management and metadata enrichment of audio visual contents.
The set up of an easy-to-use set of metadata for managing the 3D assets was, therefore, a
prominent feature. The attention was directed towards metadata standards used by digital
libraries or systems such as Europeana, the Italian Archival National System, Carare 2.0,
3D-Icons, the Dublin Core, VRA CORE or SiGECweb. The final selection of metadata built a
set deemed suitable to keep into account the peculiarities of 3D models without being too
burdensome, accounting for example for levels of detail, scientific and artistic validation or,
besides georeferencing, the height on the sea level, in order to convey information about
the correct positioning of the reconstruction with respect to the historical digital terrain
model. Besides the management of 3D assets, the DARE-IMC platform provides a viewer,
in order to appreciate the reconstruction without downloading the zip file, whenever it is
made available. The 3D models can be realised in the preferred format, but by providing
an additional .glb version, the viewer can show the reconstruction directly in the catalogue.
The API from the platform enables the use of the DARE-IMC platform contents on the DARE
Approdo website.
The approaching end of the project has led to the uniformisation of the aesthetics along
the indications of the graphic identity of the project, with interventions both on the DARE-
IMC interface and the Virtual Gallery set up, the tool available in the platform for setting up
personalised exhibitions starting queries on the audio visual contents. After the conclusion
of the project an editorial board will keep the platform and the website updated.

65
Figure 43: DARE Website https://www.darsenaravenna.it/
Figure 44: - DARE-IMC platform, the visualisation of the
3D models in the catalogue

66
COMPUTING
PROVISIONING
SERVICES

68
Bringing HPC-fueled innovation to the
EU industry
Claudio Arlandini
Cineca
More and more industries look at Cineca as
a trusted innovation partner. In our mission
to bring innovation through technologies
like HPC, AI and Big Data to industries,
particularly to Italian SMEs where the
technology gap with respect their EU
counterparts is larger, we adopted the best
practices recommended by the European
Commission for the Digital Innovation
Hubs, and we are at the center of a growing
innovation actors network (European Digital
Innovation Hubs, Competence Centres for
Industry 4.0, Industrial Associations,…).
One of the pillars of our action, together with
training, networking and access to state-
of-the-art infrastructures, is the so-called
“test-before-invest” service. Test-before-
invest means giving SMEs the unique
opportunity of experimenting with new
digital technologies – software, hardware,
business models – to understand new
opportunities and return on investments,
also including demonstration facilities and
piloting. Every innovation-introducing
projectinacompanyisassociatedwitharisk.
Trying a specific innovation in a controlled
environment, without the necessity of up-
front investments, and with the time and
effort of industry technicians co-funded by
the EU or other funding agencies means
reducing this risk. At the end, the company
will be able to make informed choices, and,
most important, without “dangerous” tying
up with specific vendors.
For Cineca, test-before-invest activities
means first of all supporting Italian
companies in EU projects-funded Proo
of Concepts (PoC) experiments, mostly
selected through competitive open calls.
The support starts well before the beginning
of a PoC. When an interesting funding
opportunity is recognized, it is disseminated
through Cineca long-standing partners, like
for instance the BI-REX Competence Centre
for Big Data, which also act as a first filter
individuating the SMEs that most best target
the opportunity. Then the SME is helped in
writing a strong proposal, an important step
since the competition is usually fierce and
SMEs rarely have internal experiences and
skill sets available for dealing with such
kind of research and innovation projects.
Winners are then supported in their PoC
experiment with computational resources
and most important the competences to
transform the tested technologies into
competitive advantages.
We will present here selected examples
from two ongoing EU projects that see
Cineca as a core partner: FF4EuroHPC and
EUHubs4Data.
FF4EuroHPC
The FF4EuroHPC Project is the third edition
of the successful Fortissimo project series.
It started at the beginning of September
2020 and will last three years. This project
is funded by the European Commission
through the Horizon 2020 Programme.
It might be considered the successor of
projects Fortissimo (July 2013 - December
2016) and Fortissimo2 (November 2015 -
December 2018).
The aim, like in the previous Fortissimo
projects, is to strengthen the global
competitiveness of European small and
medium enterprises (SMEs), by facilitating
theaccesstoandtheuseofhigh-performance
computing-related technologies. Through
two open calls, ninety-two SMEs have the
opportunity to submit innovative projects
experiments to demonstrate the benefits
of HPC cloud-based advanced modelling,
simulation and data analytics.
The project partners support the
participating SMEs in the planning,
implementation and realisation of their
proofofconcept(PoC).Whentheexperiment
is successfully concluded, a success story
raises to demonstrate the benefits of the use
of HPC technologies to those involved in the
value chain, from the end-user to the HPC-
infrastructure provider. The success stories
should inspire the industry community in
investing in R&D to develop innovative ICT
solutions for business benefits. Among the
others, these should be the main benefits for
SMEs:

69
• saving time for product/service development and decrease time to market;
• saving costs by optimising data processing tasks;
• higher levels of innovation and more patents;
• more specialised, tailor-made products and services;
• cooperative engagement between Higher Education Institutions and businesses inspires
state-of-the-art solutions;
• gaining new knowledge and know-how by R&D experts.
Cineca and its partners propelled Italian SMEs to an unprecedented success rate, with 5
selected experiments over 16 in the first open call, and 8 over 26 in the second one. These
experiments cover a wide range of technologies and industrial areas, from new materials
for aerospace to fintech and medicine. We present here a couple of noteworthy experiments.
Catching plastic waste in the Mediterranean with AI
Protecting seas and oceans against the litter is becoming a global concern and there is
a growing need worldwide for more efficient, clean and autonomous technologies to
identify and collect marine detritus, especially plastics, in a systematic and repetitive way,
combining aerial drones individuating plastic waste and marine drones to collect them.
Green Tech Solution SrL (GTS) is an agile and innovative start-up that operates in the field
of environmental protection and blue growth economy. The use of HPC makes it possible to
tackle a computational problem that GTS met during its service for recovery plastic litter in
sea: optimizing the plastic litter recovery strategy forecasting the position of hundreds of
detritus floating in the sea with suitable accuracy in space and time.
The experiment is fundamental to drive into the next phase the collaboration of the
unmanned systems as it requires >250.000 hours of deep learning which is impossible
under conventional computational systems but possible thanks to Cineca. The limited
forecasting capability of the future position of detritus thus is limiting the efficiency of
recovery of the whole system. The experiment aims to overcome this limitation and targets
to improve the current Deep Learning approach to 1. Identify and classify marine litters in
terms of dimensions and materials (PET, PPT, Biological); 2. Predict the possible trajectories
of classified waste over a longer time; 3. Search the “best” trajectory to collect as much
waste as possible under constraints.
It is expected that this system will be able to reduce the cost for public administrations to
clean up the Mediterranean coast of at least 70%, without involving the use of polluting
diesel-fueled boats.
The partners are:
• End User: Green Tech Solution SRL
• Domain Experts: Università degli Studi di Napoli Parthenope and BI-REX - Big Data Innovation & Research
Excellence
• HPC Provider: Cineca
For more info: https://www.ff4eurohpc.eu/en/experiments/2021070911193805/hpcbased_navigation_system_
for_marine_litter_hunting
Figure 45: project diagram

70
Saving children with congenital heart diseases with
HPC
Congenital heart diseases (CHDs) account for nearly one-third of all congenital birth
defects. Without the ability to alter the prevalence of CHD, interventions and resources
must be focused to improve survival and quality of life. The Modified Blalock Taussig Shunt
(mBTS) is a common palliative operation on cyanotic heart diseases, but it is associated
with significant mortality, the most threatening complications being over-shunting and
shunt thrombosis. The experiment aims at setting up an application to support surgical
planning with advanced numerical means, exposed in an interactive and effective manner.
This application uses patient-specific computationally intensive models running in a
reliable and cost-efficient way on a cloud-based HPC environment. The experiment aims
at building up an affordable decision support web application that, thanks to the medical
digital twin (MDT), allows surgeons to approach the mBTS medical intervention at best. The
tool generates the ROM of a patient-specific vascular district in which the shunt implantation
is geometrically parameterized. The ROM generates pre-defined CFD results of different
configurations of the vasculature with shunt implantation. With a dedicated User Interface,
the medical staff will be able to inspect the MDT of the patient in an interactive way, to
finalize the decisions on surgical intervention.
The partners are:
• End User: InSilicoTrials Technologies SpA
• Technology expert, ISV: RBF Morph Srl
• Application expert, End-User: Fondazione Toscana Gabriele Monasterio
• Technology expert: RINA Consulting
• HPC expert, Host Centre: Cineca
For more info:
https://www.ff4eurohpc.eu/en/experiments/2021070910512579/cloudbased_hpc_platform_to_support_
systemicpulmonary_shunting_procedures
EUHubs4Data
The objective of EUHubs4Data is to create a federation for Big Data cross-border
experimentation and innovation, providing a complete pan-European catalogue of data
Figure 46: project workflow

71
sources and services to foster data driven innovation at local and regional level. To do
this, its members, most of them being recognised as i-Spaces (DIHs) of the Big Data Value
Association, provide tools, assets and methods to European companies and society to
benefit from digital opportunities coming from data driven solutions and models. The
project aims to establish a European federation of Big Data Digital Innovation Hubs based
on the premise “global catalogue, local offer”. To achieve this, EUHubs4Data must attract,
support and engage end users from local and regional ecosystems through a common
European catalogue of federated datasets and services. EUHubs4Data brings together
relevant initiatives to create this common catalogue, capitalizing the work done by all those
initiatives, and improving the efficient use of data across the EU.
One key way to implement this vision is the selection of cross-border 12-months data-
experiments that include one European SMEs and two or more EUHubs4Data partners.
Selected through three rounds of competitive open calls.
Cineca is now supporting four experiments selected in the first two open calls (the final one
will open in September 2022). We present here one of them.
Providingnowcastingservicesofseaconditionsthanks
to HPC and AI
Off-shore industry operations security and Cities/Regions coastal area planning and
monitoring require fast and accurate sea-state forecasts. All the services now available
cover the sea areas only up to 10 km of distance from the coast. The current scenario for
sea state analysis and forecast is based on physical models and the two main operational
services over the Mediterranean Sea are run by the Copernicus Marine Environment
Monitoring Service (CMEMS) and the Italian National Agency for New Technologies, Energy
and Sustainable Economic Development (ENEA).
The experiment aims to develop an AI based model for sea-state analysis and forecast,
providing results aligned with the physical model currently used at national level, including
e.g. a sea circulation forecast daily service over the next day and a sea circulation daily
forecast over the next 5 days.
The advantages of adopting a AI-based method are:
• Much faster data processing and generation of results,
•Improvedaccuracybyensemblingvariousresults(fromdifferentinitializationparameters),
• Improved geo-spatial resolution (1x1km) and coastal coverage (up to 1 km from the coast)
also by using Earth Observation (EO) satellite data (e.g., Sentinel-1 and Sentinel-3) and
meteorological data (Mistral from Cineca),
• Wave-Circulation Nowcasting over the next 12 hours, with 30 minutes temporal resolution
and 1x1Km geo-spatial resolution,
• Easy evolution of the model and tuning to different geographic areas.
The partners are:
• End User: GMATICS srl
• Supporting DIHs: Cineca, Data Cycle Hub

72
This project has recived
foundings from the Euro-
pean High-Performance
Computing Joint Under-
taking (JU) under Grant
Agreement 951732
EuroCC
Cinzia Zannoni
Cineca
The current technology transfer project for
industry is Euro-CC (started in May 2021
and to be completed in September 2022)
and is the main project for the industry in
progress.
Thanks to European funding, the EuroCC-
Italy Competence Center, a program for
the development and promotion of High
Performance Computing tools (HPC),
advanced simulation, high performance
data analysis (HPDA) and intelligence
artificial (AI) in order to promote innovation
in industry and the public sector a national
level.
The Long-term goals of the EuroCC-Italy
Competence Center are:
• to support the national ecosystem fostering
the collaboration between research teams
and institutes, HPC computing centre and
corporates in order to support SMEs in
innovation projects.
• to become an open service platform
aggregating competences and new
institutions to support innovation
and collaboration with industry in the
continuum of HPC/HPDA/AI technologies
and methodologies.
• to attract young researchers and
professionals, support the dialogue between
generations in an innovation friendly
environment.
To reach these goals EuroCC Italy relies
on the competences and infrastructures
available in Cineca, as well as the
competences and networks of third parties,
such as Associazione Big Data, (ABD),
a scientific network in computational
sciences, BI-REX (a network of industries
and industry associations, which deliver
services for innovation capable of reaching
SMEs in the manufacturing sector;),
corporates like Eni , Dompé and Leonardo,
which have a long-lasting experience in
using and applying HPC, HPDA and AI in
their research and business processes. In
the Competence Center (CC) corporates
have the role of supporting innovations in
SMEs which belong either to their supply
and value chain, or to the green economy
national and European ecosystem, through
concrete application examples and needs.
The challenging objectives of this first
project year were:
• to set up working groups effectively
involving all third parties, monitor their
progress and ensure timing and quality of
the project deliverables.
• to develop a first level of training on HPC,
HPDA and AI through the collaboration with
third parties, which will be added to existing
training programmes for the industry;
• to set up and launch the calls for Proof-of-
Concept (PoC) projects involving SMEs and
innovative start-ups;
• to map the training activities in HPC/
HPDA/AI at the national level, to reach an
overview of the research teams working in
the field, and identify those that are already
cooperating with the industry;
• to create awareness on EuroCC-Italy and its
programme, on relevant HPC technologies,
on Quantum Computing, and on the success
stories showing business impact of HPC,
HPDA and AI based innovation, in SMEs and
corporates.
In particular, the Competence Center
intends to convey knowledge from the
world of research into targeted training
and technology transfer actions, in order to
respond to concrete needs of the national
productionfabric,thanksalsotocoordinated
intervention of trade associations.
During 2021, 11 training courses and 6
events were organized with the aim of:
- Share knowledge about HPC, HPDA, AI and
their use;
- Provide high technology skills to industry
to create innovative products;
- Create business-to-business (B2B) events
where industries could meet, locally and
nationally, organizations involved in
technology transfer linked to HPC, HPDA
and AI.
The training courses were organized face
to face in seven different cities: Bologna,
Genoa, La Spezia, Milan, Turin, Trieste and
Modena.

73
National
Recovery and
Resilience Plan

74
The EU’s long-term budget, coupled with Next Generation EU (NGEU), the temporary
instrument designed to boost the recovery, form the largest stimulus package ever financed
in Europe. A total of €2.018 trillion are helping rebuild a post-COVID-19 Europe. In this
context of extraordinary funding, Italy outlined a detailed National Recovery and Resilience
Plan – PNRR (Piano Nazionale di Ripresa e Resilienza), a large package encompassing 6
policy areas – “missions” (M1: digitization, innovation, competitiveness, culture; M2:
green revolution and ecological transition M3: infrastructure for sustainable mobility; M4:
education and research; M5: inclusion and cohesion; M6: healthcare) and 16 components.
Among the different intervention areas, of particular interest for large-scale research
infrastructure is the Mission 4 Component 2 called “from Research to business” with the
aim to strengthen and encourage the dissemination of innovative models for basic and
applied research, conducted in synergy between universities and businesses.
In this context, during 2022 Cineca has participated to several PNRR Research initiatives
that have been financed by the Italian Ministry of Research (MUR), with the aim to further
develop and exploit its large scale computational platforms:
1. National Center for High Performance Computing, Big Data and Quantum Computing
(ICSC)
2. National Biodiversity Future Center (NBFC)
3. Regional Ecosystem for Sustainable Transition in Emilia-Romagna (ECOSISTER)
4. Enlarged Partnership GRINS (Growing Resilient, INclusive and Sustainable)
5. D-3-4-Health initiative (Digital Driven Diagnostics, prognostics and therapeutics for
sustainable Health care)
1) National Research Center for HPC, Big data and Quantum
Computing (ICSC)
TheNationalResearchCenterHPC,BigdataandQuantumComputing,providesafundamental
opportunity for the national system at the scientific, industrial and economic level to face
present and future scientific and social challenges, strengthening and expanding existing
skills and infrastructural resources. The National Center will be structured according to
the hub and spoke model: the Hub is responsible for the validation and management of the
research program, whose activities are elaborated and implemented by the “Spokes” and
their affiliated institutions, as well as through open tenders. The Hub will also implement
all activities in education and training, entrepreneurship, knowledge transfer, policy and
awareness raising.
The Hub and the Spokes are made up of universities, research bodies, private and
public operators. The proposed National Center includes a transversal Spoke (Spoke 0
“Supercomputing Cloud Infrastructure”) and 10 thematic spokes.
The National Center, coordinated by National Institute of Nuclear Physics (INFN), has two
main objectives:
1) create a national computing / computing infrastructure, similar to a Datalake, grouping
the existing High Performance Computing (HPC), High Throughput Computing (HTC), Big
Data and network infrastructures, with new targeted resources, activated through internal
funding, in order to provide the scientific and industrial communities with a flexible and
uniform Cloud interface;
High Performance Computing projects
supported by the National Recovery and
Resilience Plan – Next Generation EU
Maurizio Ortali
Cineca

75
2)createanattractiveecosystemaroundtheinfrastructurethatsupportstheacademicworld
and the industrial system, favoring the exploitation of IT resources and the development of
new advanced computing technologies.
Cineca in the Center is clearly a key actor, being among the founders of the Hub, Spoke
Leader of the Infrastructure component and Affiliated party (partner) in two other Spokes
(Future HPC and Quantum Computing).
The Grant assigned to Cineca across the 3 year lifespan of the project will significantly
empower the Cineca HPC Department personnel and complement and expand new HPC
and Quantum innovative components.
Figure 47: funding members

76
2) National Biodiversity Future Center
The National Research Center (CNR) coordinates the National Biodiversity Future Center
(NBFC) with the aim of systematizing Italian research and innovation excellence on
biodiversity and delivering to Italy, after three years of work, an institution to safeguard,
develop and enhance biodiversity that is unique in its gender and sustainable over time. In
particular, the NBFC intends to carry out operational interventions aimed at:
1) Monitoring, preserving and restoring biodiversity in the marine, terrestrial and
urban ecosystems of the Peninsula. Thanks to modern technologies, expressed in terms
of Key Enabling Technologies (biotechnologies, artificial intelligence and digitization,
technologies for life sciences) it will be possible to increase knowledge relating to the
heritage of biodiversity and to map distribution, value and peculiarities in the habitats of
terrestrial ecosystems, freshwater, transitional, marine and urban. This knowledge and
technologies will be the basis of intervention programs for biodiversity conservation and
restoration also thanks to the development of increasingly realistic predictive models, the
design of new early warning technologies and the definition of tools to support functional
biodiversity and ecosystem resilience.
2) Enhance biodiversity and make it a central element on which to base sustainable
development. The study of the different levels of biodiversity and technological innovations
will make it possible to identify new products, processes (Bioprospecting) and solutions
(Nature Based Solutions - NBSs) capable on the one hand of implementing and protecting
the different living forms and on the other of using this diversity. to obtain new drugs, foods,
more efficient and sustainable materials.
These interventions will have direct and indirect effects on the well-being of the person,
for example, by reducing temperatures and pollution, increasing the tourist value and use
of degraded and highly man-made areas, favoring the relationship between people and
improving the living environment.
In the context of this National Center (project duration: 3 years), Cineca will be a key actor
being one of the founders of the Hub and will participate in the spoke 7 with the aim of
providing the Center with skills for the use of computing infrastructures and HPC.
3) ECOSISTER - Ecosystem for sustainable transition
of Emilia-Romagna
As defined in the PNRR call for proposals, the regional “Innovation Ecosystems” are
networks of state and non-state universities, Research publics, local public bodies and other
highly qualified public and private entities aimed to intervene in areas of technological
specialization consistent with the industrial and research vocations of the reference
territory, promoting and strengthening the collaboration between the research system, the
production system and local institutions.
To support his strong and highly competitive production system, coordinated by University
of Bologna and Art-ER, some Universities and Public Research Bodies of Emilia-Romagna
decided to build an integrated innovative ecosystem, in association with Universities,
Research Centers and other territorial players.
In order to maintain a leadership role in the international context and remain anchored to
the vocations that characterize the Emilia-Romagna region, the project intends to support
the ecological transition of the regional economic and social system through a process that
transversally involves all sectors, technologies and skills by combining digital transition and
sustainability with the work and well-being of people and the protection of the environment,
in line with the objectives of the Pact for Work and Climate, and integrating with regional,
national and European programs.
In the 36 months of the lifespan of the project, Cineca will participate in the activities
of Spoke 6, with a role of trait-d’union and consolidation with respect to the position of
Emilia-Romagna as National HPC Champion by aggregating its cutting-edge skills in high
performance Computing, Data Science and Technology, to support the ecological transition.

77
4) GRINS – Extended Partnership
The GRINS (Growing Resilient, INclusive and Sustainable) Consortium is an extended
partnership coordinated by University of Bologna- Department of Economics - that intends
to develop AMELIA (dAta platforM for the transfEr of knowLedge and statistIcal Analysis),
an Online Data Platform giving access to high quality data and instruments for data analysis
for a wide range of applications. It will offer tools to support fundamental and applied
research for firms and households and for policy analysis and evaluation of the actions of
public administrations. The GRINS AMELIA will create knowledge and transfer it to private
and public actors as well as to the national system of research according to the guidelines
and principles of the PNRR. The project is designed following the priorities set by the Italian
National Research Plan (PNR) and in strict adherence with the fundamental underline
goals that inspire the whole EU-NRRP action: favouring resilient, inclusive, and sustainable
growth.
In the 36 months of the lifespan of the project, Cineca will participate in the activities of
Spoke 0, with the aim to support the development of the ICT and HPC component needed
to the project, by aggregating its cutting-edge skills in high performance Computing, Data
Science and Technology.
5) D-3-4-Health Initiative
D-3-4-Health project, coordinated by Sapienza University (Rome) aims at enabling new
technologies for data collection and analysis in order to provide personalized medicine.
This approach involves the prevention, the diagnosis and therapy of oncological,
neuropsychiatric and metabolic pathologies through the creation of a “digital” and a
“biological twin” of patients.
Cineca will participate in two spokes of the project and the CINECA HPC and cloud
infrastructure will be a key asset for the proposal, as the appropriate usage of HPC
architectures has already expressed high potentiality in the field of personalized medicine
where urgent computing is critical. To this extent, CINECA can contribute leveraging the
experience acquired in several European projects with GDPR compliant procedures for
data protection and data management and supplying HPC or Cloud infrastructure to host
pseudonymized or anonymous data.
In addition to these new starting projects, Cineca has also supported the National Institute
of Oceanography and Applied Geophysics (OGS) on a PNRR proposal submitted to the MUR
in the field of “strengthening and creation of research infrastructures” (Terabit project), in
continuity with the past activities the PRACE-Italy.
Finally, Cineca has defined the contractual process and launched with the Central Institute
for the Digitization of Heritage (IC-DP) of the Ministry of Culture (MIC) the activity envisaged
for Cineca in Measure 1.1 Platforms and digital strategies for access to heritage cultural,
belonging to measure M1C3 of the PNRR: Culture and Tourism.

Report HPC

More Related Content

Similar to Report HPC (20)

More from Cineca (20)

Recently uploaded (20)

Report HPC