SlideShare a Scribd company logo
A Dramatic Shift in
Network Architecture
Presented by:
Antonella Rubicco, CEO
Emilio Billi, CTO
Stephen Ozoigbo, Chief Strategy Officer
Under NDA until February 25, 2014

© A3CUBE Inc Confidential

™
1
Company
Headquartered in the Silicon Valley
Founded in 2012 as a result of more than 5 years of research and development
operations
Assembled a very skilled and experienced A-class team in engineering and management
with proven track records and success stories
Antonella Rubicco – Founder and CEO
Emilio Billi – Founder and Chief Technology Officer
Employees – Under 25

Intellectual Property
U.S. Patents filed
#61786560 - Massive parallel petabyte scale storage system architecture
#61786537 - PCIe non-transparent bridge designed for scalability and networking enabling
the creation of complex architecture with ID based routing
#61786551 - Low-profile half length PCI Express form factor embedded PCI express
multiport switch and related accessories

© A3CUBE Inc Confidential

2
Why the need for a fundamental shift
in network architecture?
Conventional architectures to move stored data are still based on design concepts that
have been largely unchanged since the advent of storage, using single-threaded sequential
communication processes
The huge disparity between computing power and storage performance has created a
massive I/O performance gap in the enterprise as well as Big Data, HPC and data centers
The emergence of enterprise SSD technology has simply shifted the storage I/O bottleneck
from the storage device to the interconnection between storage and the CPU, exposing the
limitations of conventional PCI Express and other flash architectures
Scale-out systems today require end-to-end synchronization of metadata, or offer a limited
number of ingress/egress networking ports

© A3CUBE Inc Confidential

3
Competition from traditional network standards
Ethernet
Very well established
Many clustering topologies used
Relatively high cost (above Gigabit Ethernet)
Very high latency
High power (above Gigabit Ethernet)
Needs additional adapter cards (above Gigabit Ethernet)

Infiniband
Established in supercomputing systems
Driver model exists
Very expensive
High power
Not as widely available as Ethernet or PCIe
Requires additional adapter cards

Primarily the
competition
is from
traditional
interconnects

Standard PCIe
It’s  everywhere
It’s  the  standard  for the server I/O interface
It is not a real fabric, just a connect
It does not scale

© A3CUBE Inc Confidential

4
RONNIEE Express
A new High-Performance Cluster and data plane Interconnect based on a
disruptive pure memory mapped communication paradigm

The In Memory Network
A re-architected Network Interface Card
Merging Storage and Compute
A new ‘Brain Inspired’ Interconnection Fabric providing low latency CPU to CPU and IO to IO
traffic via a next generation Network Interface Card
A revolutionary extension of PCI Express memory mapping features
A multi-dimensional network topology with A3CUBE’s patent pending distributed nontransparent bridging (dNTB) architecture
RONNIEE Express fabric extends the standard PCIe architecture, introducing a new
architecture based on a globally shared memory container that makes clusters, expansion
boxes and shared-I/O applications easy to develop and deploy

© A3CUBE Inc Confidential

5
The New Network Fabric
Why RONNIEE Express is more powerful
than PCIe as a network fabric
RONNIEE Express is the first multi-dimensional interconnection technology designed for
exascale out storage. Scales linearly up to 64K nodes and 100s of I/O per node
Is designed to realize bottleneck-free storage in HPC, Big Data, and the datacenter for
vastly improved storage performance at vastly lower TCO
End-to-end traffic management with up to 16 million unique virtual streams between
any two endpoints
Enhanced link layer flow control with hardware based high availability (automatic
redundant links)
Detection and notification of the host manager when a link failure occurs with
containment of the error from the rest of the system
Enables revolutionary RAID-less storage architecture for hyper-scale data systems based
on  an  innovative  “In-Memory”  network  design approach

© A3CUBE Inc Confidential

6
In Memory Network
Why RONNIEE Express is different and so powerful?
The RONNIEE Express NIC uses
memory as the main
communication paradigm at
protocol level
Use PCIe direct access to memory
using  “memory  windows”  in  
combination with a globally
shared 64 bit memory address
space
A3CUBE’s  unique  NIC  architecture  
uses  these  “memory  windows”  to  
create a shared global memory
container that permits direct
communication between:
Local CPUs and remote CPUs
Memory to memory
Local and remote I/O

RONNIEE NIC

Reads and writes

server

server

server

server

server

server

Memory bridge

In Memory Network

© A3CUBE Inc Confidential

7
Brain Inspired Product Line
RONNIEE 2S
A compact PCIe-based intelligent NIC designed to maximize application performance using a
unique combination of proprietary hardware and software. The technology combines the
superior performance and ultra-low  latency  of  A3CUBE’s  In-Memory Network technology with
standard computer hardware. RONNIEE 2S Eliminates conventional communications bottlenecks
and provides multiple channels with <1µ and fast direct remote I/O connections with
nanoseconds level latency

RONNIE RIO
The first general purpose Network Interface Controller (NIC) supporting Ethernet and memory-tomemory transactions in a 3D torus topology that can plug in any server equipped with a PCIe slot.
The technology is a modern, powerful and revolutionary data fabric that presents a scalable
interconnection fabric based on a patent pending shared memory architecture that implements
the concept of distributed non-transparent bridging to massively extend PCIe features and
benefit over a complex network architecture.

RONNIEE 3
A revolutionary switch that is designed to extend the scalability of RONNIEE 2S optimized for high
performance data environments. This powerful In Memory Network provides full support for
memory-to-memory transactions without the usual software overhead achieving unmatched
efficiency and performance compared to ordinary interconnection fabrics available on the market
today

© A3CUBE Inc Confidential

8
ByOS – The Operating System for Parallel Storage
ByOS is designed to provide a Massively Parallel
Supercomputing Experience applied to data access
and computing
A new communication paradigm
Highly parallel architecture with intelligent relationship between available CPUs and I/O
Leverages supercomputing's massively parallel design concepts for exascale data access and
computing
Bottleneck-free no-metadata server architecture (Reflective memory + specific algorithms)
Designed for scaling capacity and I/O (independently)
Data access Integration with existing infrastructure (Linux , Windows, MAC OS, Unix) in minutes

© A3CUBE Inc Confidential

9
Traditional Architecture
RONNIEE and ByOS create a new massively parallel scale out storage
architecture
RONNIEE Express
Data Plane

Storage Entity

Storage Entity

Storage Entity

Storage Entity

Storage Entity

Storage Entity

Storage Entity

Storage Entity

SSDs  or  …

Storage Entity

RONNIEE Express Data
Plane

Anatomy of the
Datacenter Network

User I/O

Users
© A3CUBE Inc Confidential

10
Traditional Architecture
RONNIEE and ByOS create a new massively parallel scale out
storage architecture

DATA
PLANE

Massive Clustered Architecture

Massively
parallel data
access

USER
NETWORK
© A3CUBE Inc Confidential

11
RONNIEE Express Use Cases
Building Blocks For

Big Data / Datacenter

Biotech/Genome
Research

High-Performance
Computing

High End Routers

Computational Fluid
Dynamics

© A3CUBE Inc Confidential

Embedded
Computing

Parallel Storage

Redundant
Servers

12
RONNIEE Express new standard in performance
TCP/UDP Memory based Socket performance comparison
Standard Sockperf Benchmark
Zero copy, direct memory to
memory communication, OS
Kernel Bypass

64 byte (10 GBE ETH and PCIe) about 33 us (Ping Pong)
64 byte RONNIEE Ex 2.5us (Ping pong)

© A3CUBE Inc Confidential

13
RONNIEE Express a new standard in performance

Benchmark example applied to storage:
Unmatched IOPS
40 Terabytes of SSDs
RONNIEE Express
Multipoint direct topology (5 nodes)
Low cost SSDs
Storage server

Storage server

Memory to memory
latency (4 byte with
software 810 ns) In
any direction
Storage server

RONNIE 2S NIC multipoint direct
topology (4 Links per NIC)
Storage server

© A3CUBE Inc Confidential

Storage server

14
RONNIEE Express topology advantages
10Gbe Network

RONNIEE
Express pure 3D

RONNIEE
Express Switch

Power consumption

~ 170 KW

110 KW

75 KW

Power consumed for
cooling

60.9 KW

39.4 KW

26.8 KW

Total Power consumed

230.9 KW

149.4 KW

101.8 KW

Cost per Kwh

10 cents

10 cents

10 cents

Cost per year

$202,232

$130,850

$89,220

CO2 emitted per year

458 tons

297 tons

212 tons

Using these figures: Network Power Saving
For 10,000 nodes over a year the A3CUBE RONNIEE Express 3D solution typically saves close to
$70,000. In a five-year period this saving rises to $350,000
The RONNIEE Switch solution typically saves close to $113,000. In a five-year period this saving
rises to $565,000
Reduced power consumption also leads to a decrease in the carbon footprint. In a five-year
period the 10,000 nodes with RONNIEE 3D, when compared to a typical 10Gbe Network, lead to
a reduction of 805 tons in CO2 emissions and RONNIEE Switch lead to a reduction of 1,230
tons in CO2 emissions

© A3CUBE Inc Confidential

15
Thank You
RONNIEE Express is the new data plane massively extending PCIe capabilities and is
designed to accelerate Big Data, HPC and I/O intensive applications
Unmatched cost/performance by providing an affordable solution for
high-end applications and use cases moving HPC to HPD (High Performance
Data)
Attaining new standards of bandwidth and latency; support for an efficient IPC
paradigm, through use of non-transparent shared memory; IO virtualization in a
multi-host environment
Go To Market – OEM (server and storage), Direct, Channel

Questions?
© A3CUBE Inc Confidential

16

More Related Content

PDF
Fortissimo converged super_converged_hyper
PDF
PLX Technology Company Overview
PDF
Optimized packet processing software for networking and security
PDF
Blade Server I/O and Workloads of the Future (report)
PDF
Netapp Storage
PDF
The Role of Fibre Channel in Server Virtualization
PPT
Presentation cube
PPTX
IBM Blade University: Emulex Connects the Data Center of the Future
Fortissimo converged super_converged_hyper
PLX Technology Company Overview
Optimized packet processing software for networking and security
Blade Server I/O and Workloads of the Future (report)
Netapp Storage
The Role of Fibre Channel in Server Virtualization
Presentation cube
IBM Blade University: Emulex Connects the Data Center of the Future

What's hot (19)

PPT
Storage Area Network (San)
PPTX
Big Data Benchmarking with RDMA solutions
PPTX
Green Networking With Blade
PDF
Why Fibre Channel Fabrics to Support Flash Storage?
PPTX
Exploring the Next Wave of 10GbE with Crehan Research
PPTX
Ceph Day New York 2014: Ceph over High Performance Networks
PPTX
Ceph Day London 2014 - Ceph Over High-Performance Networks
PPTX
End User Computing with NetApp
PDF
InfiniBox z pohledu zákazníka
PPT
Cisco NetApp VMware - Long Distance VMotion
PPTX
Mellanox's Sales Strategy
PPT
Huawei Symantec Oceanspace N8000 clustered NAS Overview
PPTX
Storage Efficiency Customer Success Stories Sept 2010 power point
PDF
NetApp enterprise All Flash Storage
PPT
Storage Area networking and upcoming trends
PPTX
Storage networks
PDF
HPE Solutions for Challenges in AI and Big Data
PDF
Net App Unified Storage Architecture
PDF
Technical Report NetApp Clustered Data ONTAP 8.2: An Introduction
Storage Area Network (San)
Big Data Benchmarking with RDMA solutions
Green Networking With Blade
Why Fibre Channel Fabrics to Support Flash Storage?
Exploring the Next Wave of 10GbE with Crehan Research
Ceph Day New York 2014: Ceph over High Performance Networks
Ceph Day London 2014 - Ceph Over High-Performance Networks
End User Computing with NetApp
InfiniBox z pohledu zákazníka
Cisco NetApp VMware - Long Distance VMotion
Mellanox's Sales Strategy
Huawei Symantec Oceanspace N8000 clustered NAS Overview
Storage Efficiency Customer Success Stories Sept 2010 power point
NetApp enterprise All Flash Storage
Storage Area networking and upcoming trends
Storage networks
HPE Solutions for Challenges in AI and Big Data
Net App Unified Storage Architecture
Technical Report NetApp Clustered Data ONTAP 8.2: An Introduction
Ad

Similar to RONNIEE Express: A Dramatic Shift in Network Architecture (20)

PPTX
Disaggregation a Primer: Optimizing design for Edge Cloud & Bare Metal applic...
PPTX
SoC Solutions Enabling Server-Based Networking
PDF
Application Report: Big Data - Big Cluster Interconnects
PDF
Netronome Corporate Brochure
PDF
InTech Event | Cognitive Infrastructure for Enterprise AI
PPT
transform your busines with superior cloud economics
PDF
Introduction to NVMe Over Fabrics-V3R
PDF
RedisConf17 - Redis Enterprise on IBM Power Systems
PDF
Brocade EMC vdx connectrix 발표자료 20150212
PDF
Saviak lviv ai-2019-e-mail (1)
PPTX
Webinar: OpenEBS - Still Free and now FASTEST Kubernetes storage
PDF
OCP U.S. Summit 2017 Presentation
PDF
Netronome Corporate Brochure
PDF
Cisco UCS Servers Presentation
PDF
Cisco Connect Halifax 2018 Compute infrastructure for a hybrid cloud ucs an...
PPTX
IBM PureSystems
PPTX
Ceph Day Amsterdam 2015 - Deploying flash storage for Ceph without compromisi...
PDF
From Rack scale computers to Warehouse scale computers
PDF
Power overview 2018 08-13b
PDF
BXI: Bull eXascale Interconnect
Disaggregation a Primer: Optimizing design for Edge Cloud & Bare Metal applic...
SoC Solutions Enabling Server-Based Networking
Application Report: Big Data - Big Cluster Interconnects
Netronome Corporate Brochure
InTech Event | Cognitive Infrastructure for Enterprise AI
transform your busines with superior cloud economics
Introduction to NVMe Over Fabrics-V3R
RedisConf17 - Redis Enterprise on IBM Power Systems
Brocade EMC vdx connectrix 발표자료 20150212
Saviak lviv ai-2019-e-mail (1)
Webinar: OpenEBS - Still Free and now FASTEST Kubernetes storage
OCP U.S. Summit 2017 Presentation
Netronome Corporate Brochure
Cisco UCS Servers Presentation
Cisco Connect Halifax 2018 Compute infrastructure for a hybrid cloud ucs an...
IBM PureSystems
Ceph Day Amsterdam 2015 - Deploying flash storage for Ceph without compromisi...
From Rack scale computers to Warehouse scale computers
Power overview 2018 08-13b
BXI: Bull eXascale Interconnect
Ad

More from inside-BigData.com (20)

PDF
Major Market Shifts in IT
PDF
Preparing to program Aurora at Exascale - Early experiences and future direct...
PPTX
Transforming Private 5G Networks
PDF
The Incorporation of Machine Learning into Scientific Simulations at Lawrence...
PDF
How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...
PDF
Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ...
PDF
HPC Impact: EDA Telemetry Neural Networks
PDF
Biohybrid Robotic Jellyfish for Future Applications in Ocean Monitoring
PDF
Machine Learning for Weather Forecasts
PPTX
HPC AI Advisory Council Update
PDF
Fugaku Supercomputer joins fight against COVID-19
PDF
Energy Efficient Computing using Dynamic Tuning
PDF
HPC at Scale Enabled by DDN A3i and NVIDIA SuperPOD
PDF
State of ARM-based HPC
PDF
Versal Premium ACAP for Network and Cloud Acceleration
PDF
Zettar: Moving Massive Amounts of Data across Any Distance Efficiently
PDF
Scaling TCO in a Post Moore's Era
PDF
CUDA-Python and RAPIDS for blazing fast scientific computing
PDF
Introducing HPC with a Raspberry Pi Cluster
PDF
Overview of HPC Interconnects
Major Market Shifts in IT
Preparing to program Aurora at Exascale - Early experiences and future direct...
Transforming Private 5G Networks
The Incorporation of Machine Learning into Scientific Simulations at Lawrence...
How to Achieve High-Performance, Scalable and Distributed DNN Training on Mod...
Evolving Cyberinfrastructure, Democratizing Data, and Scaling AI to Catalyze ...
HPC Impact: EDA Telemetry Neural Networks
Biohybrid Robotic Jellyfish for Future Applications in Ocean Monitoring
Machine Learning for Weather Forecasts
HPC AI Advisory Council Update
Fugaku Supercomputer joins fight against COVID-19
Energy Efficient Computing using Dynamic Tuning
HPC at Scale Enabled by DDN A3i and NVIDIA SuperPOD
State of ARM-based HPC
Versal Premium ACAP for Network and Cloud Acceleration
Zettar: Moving Massive Amounts of Data across Any Distance Efficiently
Scaling TCO in a Post Moore's Era
CUDA-Python and RAPIDS for blazing fast scientific computing
Introducing HPC with a Raspberry Pi Cluster
Overview of HPC Interconnects

Recently uploaded (20)

PDF
NewMind AI Monthly Chronicles - July 2025
PDF
How UI/UX Design Impacts User Retention in Mobile Apps.pdf
PDF
Reach Out and Touch Someone: Haptics and Empathic Computing
PDF
Building Integrated photovoltaic BIPV_UPV.pdf
PDF
Electronic commerce courselecture one. Pdf
PDF
Unlocking AI with Model Context Protocol (MCP)
PPTX
Understanding_Digital_Forensics_Presentation.pptx
PDF
KodekX | Application Modernization Development
PDF
Review of recent advances in non-invasive hemoglobin estimation
PDF
Encapsulation theory and applications.pdf
PDF
Advanced methodologies resolving dimensionality complications for autism neur...
PDF
Encapsulation_ Review paper, used for researhc scholars
PPTX
Digital-Transformation-Roadmap-for-Companies.pptx
PDF
7 ChatGPT Prompts to Help You Define Your Ideal Customer Profile.pdf
PPTX
VMware vSphere Foundation How to Sell Presentation-Ver1.4-2-14-2024.pptx
PDF
Approach and Philosophy of On baking technology
PPTX
Effective Security Operations Center (SOC) A Modern, Strategic, and Threat-In...
PDF
Modernizing your data center with Dell and AMD
PDF
Diabetes mellitus diagnosis method based random forest with bat algorithm
PPTX
Big Data Technologies - Introduction.pptx
NewMind AI Monthly Chronicles - July 2025
How UI/UX Design Impacts User Retention in Mobile Apps.pdf
Reach Out and Touch Someone: Haptics and Empathic Computing
Building Integrated photovoltaic BIPV_UPV.pdf
Electronic commerce courselecture one. Pdf
Unlocking AI with Model Context Protocol (MCP)
Understanding_Digital_Forensics_Presentation.pptx
KodekX | Application Modernization Development
Review of recent advances in non-invasive hemoglobin estimation
Encapsulation theory and applications.pdf
Advanced methodologies resolving dimensionality complications for autism neur...
Encapsulation_ Review paper, used for researhc scholars
Digital-Transformation-Roadmap-for-Companies.pptx
7 ChatGPT Prompts to Help You Define Your Ideal Customer Profile.pdf
VMware vSphere Foundation How to Sell Presentation-Ver1.4-2-14-2024.pptx
Approach and Philosophy of On baking technology
Effective Security Operations Center (SOC) A Modern, Strategic, and Threat-In...
Modernizing your data center with Dell and AMD
Diabetes mellitus diagnosis method based random forest with bat algorithm
Big Data Technologies - Introduction.pptx

RONNIEE Express: A Dramatic Shift in Network Architecture

  • 1. A Dramatic Shift in Network Architecture Presented by: Antonella Rubicco, CEO Emilio Billi, CTO Stephen Ozoigbo, Chief Strategy Officer Under NDA until February 25, 2014 © A3CUBE Inc Confidential ™ 1
  • 2. Company Headquartered in the Silicon Valley Founded in 2012 as a result of more than 5 years of research and development operations Assembled a very skilled and experienced A-class team in engineering and management with proven track records and success stories Antonella Rubicco – Founder and CEO Emilio Billi – Founder and Chief Technology Officer Employees – Under 25 Intellectual Property U.S. Patents filed #61786560 - Massive parallel petabyte scale storage system architecture #61786537 - PCIe non-transparent bridge designed for scalability and networking enabling the creation of complex architecture with ID based routing #61786551 - Low-profile half length PCI Express form factor embedded PCI express multiport switch and related accessories © A3CUBE Inc Confidential 2
  • 3. Why the need for a fundamental shift in network architecture? Conventional architectures to move stored data are still based on design concepts that have been largely unchanged since the advent of storage, using single-threaded sequential communication processes The huge disparity between computing power and storage performance has created a massive I/O performance gap in the enterprise as well as Big Data, HPC and data centers The emergence of enterprise SSD technology has simply shifted the storage I/O bottleneck from the storage device to the interconnection between storage and the CPU, exposing the limitations of conventional PCI Express and other flash architectures Scale-out systems today require end-to-end synchronization of metadata, or offer a limited number of ingress/egress networking ports © A3CUBE Inc Confidential 3
  • 4. Competition from traditional network standards Ethernet Very well established Many clustering topologies used Relatively high cost (above Gigabit Ethernet) Very high latency High power (above Gigabit Ethernet) Needs additional adapter cards (above Gigabit Ethernet) Infiniband Established in supercomputing systems Driver model exists Very expensive High power Not as widely available as Ethernet or PCIe Requires additional adapter cards Primarily the competition is from traditional interconnects Standard PCIe It’s  everywhere It’s  the  standard  for the server I/O interface It is not a real fabric, just a connect It does not scale © A3CUBE Inc Confidential 4
  • 5. RONNIEE Express A new High-Performance Cluster and data plane Interconnect based on a disruptive pure memory mapped communication paradigm The In Memory Network A re-architected Network Interface Card Merging Storage and Compute A new ‘Brain Inspired’ Interconnection Fabric providing low latency CPU to CPU and IO to IO traffic via a next generation Network Interface Card A revolutionary extension of PCI Express memory mapping features A multi-dimensional network topology with A3CUBE’s patent pending distributed nontransparent bridging (dNTB) architecture RONNIEE Express fabric extends the standard PCIe architecture, introducing a new architecture based on a globally shared memory container that makes clusters, expansion boxes and shared-I/O applications easy to develop and deploy © A3CUBE Inc Confidential 5
  • 6. The New Network Fabric Why RONNIEE Express is more powerful than PCIe as a network fabric RONNIEE Express is the first multi-dimensional interconnection technology designed for exascale out storage. Scales linearly up to 64K nodes and 100s of I/O per node Is designed to realize bottleneck-free storage in HPC, Big Data, and the datacenter for vastly improved storage performance at vastly lower TCO End-to-end traffic management with up to 16 million unique virtual streams between any two endpoints Enhanced link layer flow control with hardware based high availability (automatic redundant links) Detection and notification of the host manager when a link failure occurs with containment of the error from the rest of the system Enables revolutionary RAID-less storage architecture for hyper-scale data systems based on  an  innovative  “In-Memory”  network  design approach © A3CUBE Inc Confidential 6
  • 7. In Memory Network Why RONNIEE Express is different and so powerful? The RONNIEE Express NIC uses memory as the main communication paradigm at protocol level Use PCIe direct access to memory using  “memory  windows”  in   combination with a globally shared 64 bit memory address space A3CUBE’s  unique  NIC  architecture   uses  these  “memory  windows”  to   create a shared global memory container that permits direct communication between: Local CPUs and remote CPUs Memory to memory Local and remote I/O RONNIEE NIC Reads and writes server server server server server server Memory bridge In Memory Network © A3CUBE Inc Confidential 7
  • 8. Brain Inspired Product Line RONNIEE 2S A compact PCIe-based intelligent NIC designed to maximize application performance using a unique combination of proprietary hardware and software. The technology combines the superior performance and ultra-low  latency  of  A3CUBE’s  In-Memory Network technology with standard computer hardware. RONNIEE 2S Eliminates conventional communications bottlenecks and provides multiple channels with <1µ and fast direct remote I/O connections with nanoseconds level latency RONNIE RIO The first general purpose Network Interface Controller (NIC) supporting Ethernet and memory-tomemory transactions in a 3D torus topology that can plug in any server equipped with a PCIe slot. The technology is a modern, powerful and revolutionary data fabric that presents a scalable interconnection fabric based on a patent pending shared memory architecture that implements the concept of distributed non-transparent bridging to massively extend PCIe features and benefit over a complex network architecture. RONNIEE 3 A revolutionary switch that is designed to extend the scalability of RONNIEE 2S optimized for high performance data environments. This powerful In Memory Network provides full support for memory-to-memory transactions without the usual software overhead achieving unmatched efficiency and performance compared to ordinary interconnection fabrics available on the market today © A3CUBE Inc Confidential 8
  • 9. ByOS – The Operating System for Parallel Storage ByOS is designed to provide a Massively Parallel Supercomputing Experience applied to data access and computing A new communication paradigm Highly parallel architecture with intelligent relationship between available CPUs and I/O Leverages supercomputing's massively parallel design concepts for exascale data access and computing Bottleneck-free no-metadata server architecture (Reflective memory + specific algorithms) Designed for scaling capacity and I/O (independently) Data access Integration with existing infrastructure (Linux , Windows, MAC OS, Unix) in minutes © A3CUBE Inc Confidential 9
  • 10. Traditional Architecture RONNIEE and ByOS create a new massively parallel scale out storage architecture RONNIEE Express Data Plane Storage Entity Storage Entity Storage Entity Storage Entity Storage Entity Storage Entity Storage Entity Storage Entity SSDs  or  … Storage Entity RONNIEE Express Data Plane Anatomy of the Datacenter Network User I/O Users © A3CUBE Inc Confidential 10
  • 11. Traditional Architecture RONNIEE and ByOS create a new massively parallel scale out storage architecture DATA PLANE Massive Clustered Architecture Massively parallel data access USER NETWORK © A3CUBE Inc Confidential 11
  • 12. RONNIEE Express Use Cases Building Blocks For Big Data / Datacenter Biotech/Genome Research High-Performance Computing High End Routers Computational Fluid Dynamics © A3CUBE Inc Confidential Embedded Computing Parallel Storage Redundant Servers 12
  • 13. RONNIEE Express new standard in performance TCP/UDP Memory based Socket performance comparison Standard Sockperf Benchmark Zero copy, direct memory to memory communication, OS Kernel Bypass 64 byte (10 GBE ETH and PCIe) about 33 us (Ping Pong) 64 byte RONNIEE Ex 2.5us (Ping pong) © A3CUBE Inc Confidential 13
  • 14. RONNIEE Express a new standard in performance Benchmark example applied to storage: Unmatched IOPS 40 Terabytes of SSDs RONNIEE Express Multipoint direct topology (5 nodes) Low cost SSDs Storage server Storage server Memory to memory latency (4 byte with software 810 ns) In any direction Storage server RONNIE 2S NIC multipoint direct topology (4 Links per NIC) Storage server © A3CUBE Inc Confidential Storage server 14
  • 15. RONNIEE Express topology advantages 10Gbe Network RONNIEE Express pure 3D RONNIEE Express Switch Power consumption ~ 170 KW 110 KW 75 KW Power consumed for cooling 60.9 KW 39.4 KW 26.8 KW Total Power consumed 230.9 KW 149.4 KW 101.8 KW Cost per Kwh 10 cents 10 cents 10 cents Cost per year $202,232 $130,850 $89,220 CO2 emitted per year 458 tons 297 tons 212 tons Using these figures: Network Power Saving For 10,000 nodes over a year the A3CUBE RONNIEE Express 3D solution typically saves close to $70,000. In a five-year period this saving rises to $350,000 The RONNIEE Switch solution typically saves close to $113,000. In a five-year period this saving rises to $565,000 Reduced power consumption also leads to a decrease in the carbon footprint. In a five-year period the 10,000 nodes with RONNIEE 3D, when compared to a typical 10Gbe Network, lead to a reduction of 805 tons in CO2 emissions and RONNIEE Switch lead to a reduction of 1,230 tons in CO2 emissions © A3CUBE Inc Confidential 15
  • 16. Thank You RONNIEE Express is the new data plane massively extending PCIe capabilities and is designed to accelerate Big Data, HPC and I/O intensive applications Unmatched cost/performance by providing an affordable solution for high-end applications and use cases moving HPC to HPD (High Performance Data) Attaining new standards of bandwidth and latency; support for an efficient IPC paradigm, through use of non-transparent shared memory; IO virtualization in a multi-host environment Go To Market – OEM (server and storage), Direct, Channel Questions? © A3CUBE Inc Confidential 16