Intel® Xeon Phi™ processor (codenamed Knights Landing)applications Code and Recipe Guide

Lammps Coarse-Grain Water Simulation
LAMMPS is a classical molecular dynamics code, and an acronym for
Large-scale Atomic/Molecular Massively Parallel Simulator. It is used to
simulate the movement of atoms to develop better therapeutics, improve
alternative energy devices, develop new materials, and more.
Application: Coarse-grain water simulation with LAMMPS using Stillinger-
Weber potential. More at http://lammps.sandia.gov/
Code: In main LAMMPS repository. Recipe: Available here
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems,
components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated
purchases, including the performance of that product when combined with other products. For more information go to http://www.intel.com/performance *Other names and brands may be claimed as the property of others
Life Sciences
Ice formation in water droplet
wetting a flat surface with
coarse-grain model in LAMMPS
Image Source: Comput. Phys. Commun., 2013, 184, 2785-2793
2

"LAMMPS is an established code in the molecular dynamics simulation community with users from a variety of
disciplines including materials modeling, biology, and many others. Over the past twenty years, the code base
has been refactored and extended with high performing kernels and more complex interatomic and coarse-
grained potentials.
In partnership with Intel, software optimizations focused on several commonly-used LAMMPS models have
enabled simulations on the latest Intel® processors to now run up to 7.6X faster with over 9X the energy
efficiency, compared to LAMMPS runs a year ago on previous Intel processors. Furthermore, using a single
Intel® Xeon Phi™ processor (codenamed Knights Landing), the optimized LAMMPS code now runs up to
1.95X faster with 2.83X better performance per watt, when compared to two Intel® Xeon® E5 v3 processors.
These achievements are enabling the LAMMPS user community to overcome barriers in computational
modeling, enabling new research with larger simulation sizes and longer timescales." Steve Plimpton,
Distinguished Member of Technical Staff, Sandia National Laboratories.”
- Steve Plimpton, Distinguished Member of the Technical Staff,
Sandia National Laboratories
See the LAMMPS Technology Brief here
3

vasp*
The Vienna Ab initio Simulation Package (VASP) is a computer program for
atomic scale materials modeling and performs electronic structure calculations
and quantum-mechanical molecular dynamics from first principles.
Application: Vienna Ab initio Simulation Package (VASP)
Code: Available here Recipe: See configuration details. Also, check for future
availability here
Power Data: Total system wall power is measured out-of-band over iPMI
interface, polling the BMC chip every half second. Energy usage is matched to
internally timed code segment to arrive at performance-per-watt estimate.
Image Source: Intel
Material Sciences
purchases, including the performance of that product when combined with other products. For more information go to http://www.intel.com/performance *Other names and brands may be claimed as the property of others 4

Intel embree V2.13
Embree is a collection of high-performance ray tracing kernels, developed at
Intel. The target user of Embree are graphics application engineers that
want to improve the performance of their application by leveraging the
optimized ray tracing kernels of Embree. Embree supports runtime code
selection to choose the traversal and build algorithms that best matches the
instruction set of the CPU.
Application: Path tracer renderer using Embree v2.13, ICC 2016 Update 1,
Intel® SPMD program compiler (Intel® ISPC) 1.9.1, and OptiX* RT libraries.
Code: Available here Recipe: Available here
Image Source:
Intel
DCC/Visualization
5

Industrial standard benchmark that uses Monte Carlo method for pricing
European call options. It pre-generates random numbers then uses them in
all options pricing processes. Used by all financial firms to price derivatives
with multiple dimensions. Uses the stock price, strike price and time as
input streams then creates a call output stream.
Application: Monte Carlo European Options
Code and Recipe: Available here
Monte Carlo European options
benchmark*
Financial Services
Image Source: US gov.
6

Black-Scholes benchmark*
Industrial standard benchmark that calculates call and put option price
using the Black-Scholes-Merton Formula. Used by all financial firms to
price derivatives with multiple dimensions. Stock price, strike price and time
are input streams that create call and put as output streams.
Application: Black-Scholes formula
Financial Services
Image Source: US gov

STAC-A2* BENCHMARK
The STAC-A2 Benchmark suite is the industry standard created by the
financial community to test technology stacks used for compute-intensive
analytic workloads involved in pricing and risk management
Application: Intel Composer XE STAC Pack Rev. H
Image Source: Intel
“STAC” and all STAC names are
trademarks or registered trademarks
of the Securities Technology
Analysis Center LLC.
Financial Services

Binomial options Pricing
benchmark*
Binomial Tree Option pricing method is an industrial standard
benchmark that calculates call option prices. Used by financial firms
especially for options that involve early exercise clause. Uses stock
price, strike price and time as input streams, then creates a call
output stream.
Application: Binomial Tree Option Pricing
Code and Recipe: Check for availability here
Financial Services
Image Source: Science Direct
9

CP2K is a powerful and scalable program for atomistic simulations of a wide
range of systems, including condensed phase, molecular systems and complex
interfaces. CP2K features a wide range of atomistic interaction models including
classical potentials, semi-empirical schemes, Density Functional Theory (DFT),
Hartree-Fock (HF), and post-HF correlation methods such as MP2 and RPA. The
program was a Gordon Bell Finalist in 2015. CP2K is freely available.
Application: CP2K Quantum Chemistry & Solid State Physics Software Package
Code: Available Here Recipe: Check for availability here
Courtesy of ETH Zurich. (url)
Material Sciences Linear Scaling (LS) Density
Function Theory (DFT)

The MILC Code is used to study quantum chromodynamics (QCD), the theory
of the strong interactions of subatomic physics and is written by the MIMD
Lattice Computation (MILC) collaboration.
Application: Trinity MILC provided by NERSC as part of Trinity8 suite
Code: Original NERSC Benchmark code is here; contact Intel for Optimized
Code Recipe: Check for availability here
MILC*Physics - QCD
Image Credit: Brookhaven Lab (BNL)

QPhiX is an optimized solver library for QCD on Intel® Xeon® and Intel® Xeon
Phi™ processors and provides implementation for Dslash operator and CG,
BICGStab and mixed precision solvers for Wilson and Clover improved Wilson
Quarks.
Application: QPhiX Test Benchmark (time_dslash_noqdp), QUDA* (NVIDIA*)
Code: https://github.com/JeffersonLab/qphix Recipe: Follow the instructions in
the download package
QPhix*Physics - QCD
Flux tubes between 3 quarks.
Credit: Dr Derek Leinweber, Centre for the Subatomic structure of Matter
(CSSM) and Department of Physics, University of Adelaide, 5005
Australia

BAW American options
Approximation benchmark*
Popular analytical method of pricing exchange-traded American
options using quadratic approximation. Uses an underlying asset and
carrying cost rate as key inputs and prices commodity options,
futures and foreign exchange options, precious metals, long-terms
debt and stock indexes with continuous dividend yields. Uses the
stock price, strike price and time as input streams and creates a call
output stream.
Application: Barone-Adesi and Whaley (BAW) American Options
Approximation
Financial Services
Image Sources: Parsiad.azimzadeh, finance.bi.no
13

Monte Carlo European options
benchmark*
Industrial standard benchmark that uses Monte Carlo method for
pricing European call options. It pre-generates random numbers then
uses them in all options pricing processes. Used by all financial firms
to price derivatives with multiple dimensions. Uses the stock price,
strike price and time as input streams then creates a call output
stream.
Application: Monte Carlo European Options
Financial Services
14

Black-Scholes benchmark*
Industrial standard benchmark that calculates call and put option price
using the Black-Scholes-Merton Formula. Used by all financial firms to
price derivatives with multiple dimensions. Stock price, strike price and time
are input streams that create call and put as output streams.
Application: Black-Scholes formula
Financial Services
Image Source: US gov
15

Binomial options Pricing
benchmark*
Binomial Tree Option pricing method is an industrial standard
benchmark that calculates call option prices. Used by financial firms
especially for options that involve early exercise clause. Uses stock
price, strike price and time as input streams, then creates a call
output stream.
Application: Binomial Tree Option Pricing
Code and Recipe: Check for availability here
Financial Services
Image Source: Science Direct
16

Amber16* Generalized Born
(Implicit Solvent)
Amber* is a bio related simulation code for DNA, RNA, protein, and other
bio-molecules. Amber has two solvers: Particle Mesh Ewald (PME),
known as Explicit, and Generalized Born (GB), known as Implicit. Amber
is written in Fortran 90 and is mainly MPI*, OpenMP* and Vectorization
parallelized.
Application: Amber 16 PMEMD Implicit
Code: In main Amber GIT repository. Recipe: http://ambermd.org/intel
Life Sciences
Images Source: Intel
17

Amber16* PME (explicit solvent)
Amber* is a bio related simulation code for DNA, RNA, protein, and other bio-
molecules. Amber has two solvers: Particle Mesh Ewald (PME), known as
Explicit, and Generalized Born (GB), known as Implicit. Amber is written in
Fortran 90 and is mainly MPI*, OpenMP* and Vectorization parallelized.
Application: Amber 16 PMEMD Explicit
Code: In main Amber GIT repository. Recipe: http://ambermd.org/intel
Life Sciences
Images Source: Intel
18

NWChem NWPW AIMD
SIMULATION
NWChem is a computational chemistry software package which also includes
quantum chemical and molecular dynamics functionality. It was designed to
run on high-performance parallel supercomputers as well as conventional
workstation clusters. It aims to be scalable both in its ability to treat large
problems efficiently, and in its usage of available parallel computing
resources.
Application: Ab-initio molecular dynamics simulation (NWChem NWPW),
water128 benchmark
Code: Main NWChem repository at https://svn.pnl.gov/svn/nwchem/trunk,
SVN rev 28860. Recipe: Check for availability here
Life Sciences
19
Images Source: US Govt.; NWChem

ROME/sml
ROME (Refinement and Optimization via Machine lEarning for cryo-EM) is one
of the major research software packages from the Dana-Farber Cancer Institute.
ROME is a parallel computing software system dedicated to high-resolution cryo-
EM structure determination and data analysis, implementing advanced machine
learning approaches optimized for HPC clusters. ROME 1.0 introduces SML
(statistical manifold learning)-based deep classification, following MAP-based
(maximum a posteriori) image alignment.
Application: ROME/SML
Images Source: L. Zhang, S. Chen, J. Ruan, J. Wu, A.B. Tong, Q. Yin, Y. Li, L. David, A. Lu, W.L.
Wang, C. Marks, Q. Ouyang, X. Zhang, Y. Mao*, H. Wu*. Cryo-EM structure of the activated NAIP2-
NLRC4 inflammasome reveals nucleated polymerization. Science 350, 404-409 (2015).
Life Sciences
20

RELION (REgularised LIkelihood OptimisatioN) is a stand-alone computer
program that employs an empirical Bayesian approach to refinement of 3D
reconstructions or 2D class averages in Cryo-EM.
Application: RELION 1.4
RELIONLife Sciences
Images Source: http://www.rcsb.org/
21

Nanoscale Molecular Dynamics
program*
Nanoscale Molecular Dynamics program (NAMD) is a parallel molecular
dynamics code designed for high-performance simulation of large
biomolecular systems. Based on Charm++ parallel objects, NAMD scales to
hundreds of cores for typical simulations and beyond 200,000 cores for the
largest simulations.
Application: NAMD 2.11
Code: http://www.ks.uiuc.edu/Research/namd/
Recipe: Check for availability here
Life Sciences
Image Source: Use approved by NAMD
22

GROMACS*
GROMACS (GROningen MAchine for Chemical Simulations ) is a
versatile package to perform classical Molecular Dynamics simulations.
Heavily optimized for most modern platforms and provides extremely
high performance compared to all other MD codes.
Application: GROMACS
Code: Available here
Recipe: All optimizations merged in GROMACS 2016 branch, MKL FFT
Life Sciences
Images Source: Used with permission
23

GROMACS*
GROMACS (GROningen MAchine for Chemical Simulations ) is a
versatile package to perform classical Molecular Dynamics
simulations. Heavily optimized for most modern platforms and
provides extremely high performance compared to all other MD
codes.
Application: GROMACS (Intel® AVX-512 speedup)
Recipe: All optimizations merged in GROMACS 2016 branch, MKL FFT
Life Sciences
Images Source: Used with permission
24

LB3D is a 3D Lattice Boltzmann method kernel developed by Carlos Rosales
of Texas Advanced Computer Center (TACC) and used in multiphase flows
with applications in the multiphase reactors and separation systems.
Application: LBS3D (Advanced support for multiphase flows with large
density and viscosity ratios)
Recipe: No code changes were required. Recompile with Intel® AVX-512.
Texas advanced computer center
LBS3D*
Manufacturing
Image Source: Used with permission
25

Openfoam*
OpenFOAM (for "Open source Field Operation And Manipulation")
is a C++ toolbox for the development of customized numerical
solvers, and pre-/post-processing utilities for the solution of
continuum mechanics problems, including computational fluid
dynamics (CFD).
Application: OpenFOAM
Optimizations: https://github.com/OpenFOAM/OpenFOAM-Intel
Image Source: Intel
Manufacturing
26

NASA Overflow*
OVERFLOW is a 3D time marching implicit Navier-Stokes computational
fluid dynamics simulator developed by NASA and used across aerospace
and other industries.
Application: OVERFLOW 2.2L has an extensive feature set supporting
collision detection and modelling with support for thin layer and full viscous
terms.
Code: http://overflow.larc.nasa.gov/
Recipe: No code changes were required. Recompile with Intel® AVX-512.
Manufacturing
UH-60 Black Hawk helicopter and Navier-Stokes detached eddy simulation of a flexible
UH-60 rotor using the OVERFLOW CFD code in forward flight.
Image Source: Neal Chaderjian and Tim Sandstrom, NASA/Ames
27

LS-DYNA is an advanced general-purpose Multiphysics simulation software
package developed by LSTC. While the package continues to contain more
and more possibilities for the calculation of many complex, real world
problems, its origins and core-competency lie in highly nonlinear transient
dynamic finite element analysis (FEA) using explicit time integration. LS-
DYNA is used by the automobile, aerospace, construction, military,
manufacturing, and bioengineering industries.
28
Lstc ls-dyna*Manufacturing
Application:
LS-DYNA non-linear
explicit FEA software
Code: Available here.
Recipe NA
RELATED
VIDEO

OpenLB*
The OpenLB project provides a C++ package for the implementation of
lattice Boltzmann simulations that is general enough to address a vast
range of problems in computational fluid dynamics. The package is mainly
intended as a programming support for researchers and engineers who
simulate fluid flows by means of a Lattice Boltzmann method.
Application: OpenLB
Manufacturing
29

High resolution flow solver on unstructured
meshes (hIfun)*
HiFUN is a general purpose flow solver employing unstructured data based
algorithms and fine tuned to solve typical aerospace applications. The code
has been extensively used for solving a number of problems, over a wide
range of Mach numbers, ranging from airship aerodynamics to aerodynamics
of hypersonic vehicles.
Application: HiFUN
Baseline Code: Proprietary code (2.5.1 beta version), http://sandi.co.in/
Optimized Code: Proprietary code-Pending check-in
Recipe: “mpiifort” OPTIONS=“-xMIC-AVX512 –O3”
Manufacturing
Image Source: SandI Engineering
30

GE TACOMA*
TACOMA is an explicit Navier-Stokes computational fluid dynamics
software with structured and unstructured grid capability developed by
General Electric*. It has an extensive feature set including multi-grid
methods and is routinely used for the design of turbomachinary at GE.
Application: TACOMA
Code: GE internal in-house proprietary code
Recipe: Code optimized for outer-loop vectorization of key routines
Manufacturing
Left: Canonical compressor simulation, Right: Low pressure turbine simulation
Image Source: Brian Mitchell, GE Global Research
31

Nucleus for European Modelling
of the Ocean* (NEMO)
Nucleus for European Modelling of the Ocean (NEMO) is an ocean modelling
framework composed of "engines" in an "environment". The "engines" provide
numerical solutions of ocean, sea-ice, tracers and biochemistry equations and
their related physics. The "environment" consists of the pre- and post-processing
tools, the interface to the other components of the Earth System, the user
interface, the computer dependent functions and the documentation of the
system. NEMO allows several ocean related components of the earth system to
work together or separately.
Application: NEMO 3.6
Code: http://www.nemo-ocean.eu/
Recipe: See configuration details or check availability here
Climate &
Weather
Image Source: Sediment Dynamics in the Black Sea
32

Danish Meteorological Institute
HIROMB-BOOS-Model*
The Danish Meteorological Institute (DMI) institute was founded to make
observations, communicate them to the general public, and to develop
scientific meteorology.
Application: DMI HIROMB-BOOS-Model is a 3D ocean circulation model
code forced by atmospheric meteo-fields from a weather model.
Code: To get access to the code and test cases, please contact DMI
Recipe: Check for availability here
Climate &
Weather
Image Source: DMI
33

Non-hydrostatic Icosahedral Model (NIM) is developed by NOAA’s Earth
System Research Laboratory. NIM is used for earth system modeling and
weather and climate prediction. NIM is a multi-scale model designed to
improve tropical convective clouds and to extend weather forecasts into
intra-seasonal predictions.
Application: Non-hydrostatic Icosahedral Model
Code: To get access to the code and test cases, please contact NOAA
Recipe: Build and run instructions are similar to those for Intel® Xeon®
processor E5-2697 v4, except build with –xMIC-AVX512 flag and run with
“numactl –m 1” prepended to the command-line.
Non-hydrostatic Icosahedral
Model * (nim)
Climate &
Weather
Image Source: NOAA
34

MPAS Ocean 4.0*
MPAS (Model for Prediction Across Scales) is a suite of programs for
atmosphere, ocean, and other earth-system simulation. LANL is
primarily responsible for the MPAS Ocean (MPAS-O) model. MPAS-O has
demonstrated the ability to accurately reproduce mesoscale activity.
(workload contact: Doug Jacobson, LANL,
jacobsen.douglas@gmail.com)
Application: MPAS-O
Code: Available here Recipe: See the configuration details slide
Image Source: Los Alamos
National Laboratory
Climate &
Weather
35

HOMME is the spectral element dynamical core that solves the equations of
motion in the CAM5 atmospheric model, part of the Community Earth System
Model (CESM) jointly developed by NSF and DOE.
Application: Baroclinic instability simulation in a “whole atmosphere” (extending to
lower thermosphere) configuration.
Code: Request access to development branches here.
Recipe: cmake -DADD_Fortran_FLAGS="-O3 -xMIC-AVX512 -fp-model fast"
HOMME atmospheric dynamical core *
Numerical climate Simulation
Climate &
Weather
Image source: Used by
permission.
Global 1/8 degree
simulation run using
CAM5 with the HOMME
spectral element
dynamical core.
Visualization from
http://www.sdav-
scidac.org/29-
highlights/visualization/64-
precipitable-water-vis.html.
36

GNAQPMS*
GNAQPMS is an in-house code from the Institute of Atmospheric Physics (IAP),
China Academy of Sciences (CAS), parallelized with hybrid MPI+OpenMP and
written in Fortran. The application is the global multi-scale chemistry transport
model which can simulate the trace gases including ozone, NOx , CO and main
aerosols including dusts, sea salts, BC, OC, sulfate and nitrate in multi-special
resolutions.
Application: GNAQPMS.
Code: In-house code. To access, please contact tangxiao@mail.iap.ac.cn
Recipe: Please contact tangxiao@mail.iap.ac.cn
Image Source: PRC IAP
Climate &
Weather
37

Weather & Research Forecast Model*
Numerical Weather Simulation
The WRF Model is a numerical weather prediction system designed to serve
atmospheric research and operational forecasting needs. Currently in
operational use at NCEP, AFWA, NASA, NOAA, etc.
Application: The Weather & Research Forecast Model* (WRF) WRFV3.6.1
Conus12km. Community code is managed by NCAR. CONUS12KM
benchmark is an adhoc industry standard workload and is widely cited.
Code: Available here. (WRF 3.6 & 3.6.1) Recipe: Select Intel MIC
configuration on build. Check for availability here
Climate &
Weather
Image Source: NOAA
38

Integrated Forecasting System (IFS) is a comprehensive Earth system
modelling software developed by The European Centre for Medium-Range
Weather Forecasts (ECMWF) in collaboration with Météo-France. IFS forms
the basis of all data assimilation and numerical weather prediction activities at
ECMWF.
IFS RAPS14*
Climate &
Weather
Application: IFS RAPS14
Code and Recipe: IFS RAPS14 code
is proprietary and available to
consortium members under license.
Minor source code changes were
applied to some of the key routines to
ensure vectorization. See the
configuration details for code and
recipe information. Check for future
availability here.
Image Source: ECMWF
39

Parallel Ocean Program*
Parallel Ocean Program (POP) is an ocean circulation model that solves the three-
dimensional primitive equations for ocean dynamics. It consists of the baroclinic
and barotropic solvers that solve 3-D equations explicitly and 2-D surface pressure
implicitly, respectively. It is widely used for oceanography. The code is Open
Source.
Application: POP (Parallel Ocean Program).
Code: http://www2.cesm.ucar.edu/
Recipe: Available here
Climate &
Weather
Image Source: FIO, China
40

MASNUM WAVE*
MASNUM WAVE model is the 3rd generation surface wave model proposed
early in 1990s in LAGFD (Laboratory of Geophysical Fluid Dynamics) from
FIO. The application is used to simulate and predict the wave process by
solving the wave energy spectrum balance equation and its complicated
characteristic equations in wave-number space, which is written in Fortran
and parallelized with MPI+OMP.
Application: MASNUM WAVE.
Image Source: FIO, China
Climate &
Weather
41

Trinity Benchmarks - Optimized
cluster (GFLOPS)
Trinity is a set of benchmark programs used as part of the joint NERSC/ACES
NERSC-8/Trinity system procurement.
Code: In main NERSC website. Recipe: Check for availability here
 AMG: Parallel algebraic multigrid solver for linear systems
 MiniFE: Finite Element mini-app
 UMT: 3D, deterministic, multigroup, photon transport code for unstructured
meshes
 SNAP: proxy application to model the performance of a modern discrete
ordinates neutral particle transport application
 GTC: Gyrokinetic Particle Simulation of Turbulent Transport in Burning
Plasmas
 MILC: MIMD Lattice Computation (MILC) collaboration kernel used to study
quantum chromodynamics (QCD)
 MiniGhost: Finite Difference mini-app
Material Sciences
42

Trinity Benchmarks - Single Node
Optimized (GFLOPS)
Trinity is a set of benchmark programs used as part of the joint NERSC/ACES
NERSC-8/Trinity system procurement.
Code: In main NERSC website. Recipe: Check for availability here
 AMG: Parallel algebraic multigrid solver for linear systems
 MiniFE: Finite Element mini-app
 UMT: 3D, deterministic, multigroup, photon transport code for unstructured
meshes
 SNAP: proxy application to model the performance of a modern discrete
ordinates neutral particle transport application
 GTC: Gyrokinetic Particle Simulation of Turbulent Transport in Burning
Plasmas
 MILC: MIMD Lattice Computation (MILC) collaboration kernel used to study
quantum chromodynamics (QCD)
 MiniGhost: Finite Difference mini-app
Material Sciences
43

CP2K is a powerful and scalable program for atomistic simulations of a wide range of
systems, including condensed phase, molecular systems and complex interfaces.
CP2K features a wide range of atomistic interaction models including classical
potentials, semi-empirical schemes, Density Functional Theory (DFT), Hartree-Fock
(HF), and post-HF correlation methods such as MP2 and RPA. The program was a
Gordon Bell Finalist in 2015. CP2K is freely available.
Workload: Linear Scaling DFT used to compute the electronic structure of an
amorphous material with 14K atoms using the DZVP-MOLOPT basis set.
Courtesy of ETH Zurich. (url)
Material Sciences Linear Scaling Density Function
Theory
44

CP2K is a powerful and scalable program for atomistic simulations of a wide range of
systems, including condensed phase, molecular systems and complex interfaces.
CP2K features a wide range of atomistic interaction models including classical
potentials, semi-empirical schemes, Density Functional Theory (DFT), Hartree-Fock
(HF), and post-HF correlation methods such as MP2 and RPA. The program was a
Gordon Bell Finalist in 2015. CP2K is freely available.
Workload: Random Phase Approximation (RPA) energies using hybrid XC functionals
with exact Hartree-Fock Exchange (HFX).
Material Sciences Random Phase Approximation
(RPA using exact HFX)
45

BerkeleyGW Package is a set of computer codes that calculates the
quasiparticle properties and the optical responses of a large variety of
materials from bulk periodic crystals to nanostructures such as slabs, wires
and molecules. It is a massively parallel computational package for electron
excited state properties that is based on many-body perturbation theory
employing the ab initio GW and GW plus Bethe-Salpeter equation
methodology. Sigma is the second half of the GW code. It gives the
quasiparticle self-energies and dispersion relation for quasielectron and
quasihole states.
Application: BerkeleyGW Sigma phase of Benzene analysis
Code: http://www.berkeleygw.org
Deslippe, Jack, Georgy Samsonidze, David A. Strubbe, Manish Jain, Marvin L. Cohen, and
Steven G. Louie. "BerkeleyGW: A massively parallel computer package for the calculation
of the quasiparticle and optical properties of materials and nanostructures." Computer
Physics Communications 183, no. 6 (2012): 1269-1289.
Recipe: -xMIC-AVX512 -Ofast –qopenmp
Material Sciences berkeleygw*
46

PWmat is a commercial software using plane wave pseudopotential method for
density functional theory (DFT) material simulations. It is an ab initio code,
meaning it uses initial atomic positions to predict the material properties. PWmat is
developed and optimized by Beijing LongXun Inc. based on the open source code
PEtot (PEtot has a BSD 3-Clause license).
Application: PWmat
Code: Commercial software. Contact LongXun Inc. (support@pwmat.com). PEtot
can be downloaded as a reference (http://cmsn.lbl.gov/html/PEtot/PEtot.html).
Recipe: Evaluation binaries: http://www.pwmat.com/pwmat_performance
(encrypted). Contact LongXun Inc. (support@pwmat.com) for approval at first.
PWmat*Material Sciences
)( o
io
i
i
ii PPAP





ij
jiii PPP
,1
|
iiiii P  sincos 


ij
jiii 
Precond.CG step
Projection *
Orth *
Sub_diag *
Line minimize
nline
Hpsi *
ji Hjih ),( Sub_diag *
iiii HP   Hpsi *
MPI_Send & Recv (wf)
Nonlocal & FFT
MPI_Allreduce (mx)
Zgemm
iiii HP  
Zpotrf & Zgemm
MPI_Allreduce(mx)
Zheev & Zgemm
MPI_Allreduce(mx)
Zheev & Zgemm
MPI_Allreduce(mx)
MPI_Send & Recv (wf)
Nonlocal & FFT
ji Hjih ),(
Image Source: LongXun Inc.
47

Quantum EsPresso*
AUSURF112 (PWSCF)
Quantum ESPRESSO is an integrated suite of Open-Source computer codes
for electronic-structure calculations and materials modeling at the nanoscale.
It is based on density-functional theory, plane waves, and pseudopotentials.
Application: Quantum ESPRESSO AUSURF112 (QE 5.4)
Recipe: See the configuration details slide
Image Source: Argonne National Laboratory
Material Sciences
48

Quantum EsPresso* PRACE
GRIR443 (PWSCF)
Application: Quantum ESPRESSO 6.0
Code: Available here. Recipe: See the configuration details slide
Material Sciences
49

Quantum EsPresso* PRACE
Test Case A (CP)
Application: Quantum ESPRESSO 6.0
Code: Available here. Recipe: See the configuration details slide.
Material Sciences
50

QPhiX is an optimized solver library for QCD on Intel® Xeon® processors and
Intel® Xeon Phi™ processors and provides implementation for Dslash operator
and CG, BICGStab and mixed precision solvers for Wilson and Clover improved
Wilson Quarks.
Application: QPhiX Test Benchmark (time_dslash_noqdp)
Code: Available here Recipe: Follow the instructions in the download package
QPHIX*Physics - QCD
Flux tubes between 3 quarks.
Credit: Dr Derek Leinweber,
Centre for the Subatomic
structure of Matter (CSSM)
and Department of Physics,
University of Adelaide, 5005
Australia
51

The MILC Code is used to study quantum chromodynamics (QCD), the theory
of the strong interactions of subatomic physics and is written by the MIMD
Lattice Computation (MILC) collaboration.
Application: Trinity MILC provided by NERSC as part of Trinity8 suite
Code: Original NERSC Benchmark code is here; contact Intel for Optimized
Code Recipe: Check for availability here
MILC*Physics - QCD
52

Cloverleaf*
The CloverLeaf* code investigates the behavior of fluids under high
temperatures and pressures, which potentially cause shock fronts to form. It is
common for hydrocodes to be constructed using one of two formulations –
Lagrangian, in which a mesh is constructed and evolved through time, or
Eulerian, where material flow is calculated relative to a fixed spatial grid.
Application: CloverLeaf
Code: https://github.com/UK-MAC/CloverLeaf
Recipe: make COMPILER=INTEL MPI_COMPILER=mpiifort
C_MPI_COMPILER=mpiicc OPTIONS=“-xMIC-AVX512” C_OPTIONS=“-
xMIC-AVX512”
Physics - Hydrodynamics
Image Source: Intel
53

Chroma*
The Chroma package supports data-parallel programming constructs for lattice
field theory and in particular lattice QCD. It uses the SciDAC QDP++ data-parallel
programming (in C++) that presents a single high-level code image to the user,
but can generate highly optimized code for many architectural systems including
single node workstations, multi and many-core nodes, clusters of nodes via QMP,
and classic vector computers.
Application: Chroma “hmc”
Code: Available here Recipe: Check for availability here
Physics - QCD
54

Soft sphere simulation*
Soft Sphere is a 3D Molecular Dynamic simulation of IPE-CAS (Institute of Process
Engineering), China. Using sphere particles to simulate structured molecules and
calculation based on BKS (Beest-Kramers-Santen) Experience Potential Model
allows scientists to fight epidemics like the 2014 Ebola virus.
Application: Soft Sphere Simulation
The simulation of force-driven flow in the nano-scale channel
Image Source: IPE-CAS, China
begin
end
Physics
55

Petsc – Portable, extensible toolkit for
scientific computation
PETSc – the Portable, Extensible Toolkit for Scientific Computation – is a
suite of data structures and routines for the scalable (parallel) solution of
scientific applications modeled by partial differential equations.
Application: Solution of the incompressible, variable viscosity Stokes
equation in 3d using Q1Q1 elements, using a state-of-the art Schur
complement-based approach robust to large viscosity jumps.
Code: PETSc development code is completely open and available here.
Recipe: “-g -O3 -fp-model fast” and “-xMIC-AVX512” or “-xCORE-AVX2
Image Source: Peter Lichtner, OFM Research
Physics
56

YASK HPC Stencils, AWP-ODC
kernel
YASK, Yet Another Stencil Kernel, is a framework to facilitate design
exploration and tuning of HPC kernels. One of the stencils included in YASK
is awp-odc, a staggered-grid finite difference scheme used to approximate
the 3D velocity-stress elastodynamic equations:
http://hpgeoc.sdsc.edu/AWPODC. Applications using this stencil simulate the
effect of earthquakes to help evaluate designs for buildings and other at-risk
structures.
Application: YASK, AWP stencil (a single-node proxy for the compute kernel
in the Anelastic Wave Propagation application)
Geophysics
Image Source: USGS
57

YASK HPC Stencils, iso3dfd
kernel
YASK, Yet Another Stencil Kernel, is a framework to facilitate design
exploration and tuning of HPC kernels. One of the stencils included in YASK
is iso3dfd, a finite-difference code found in seismic imaging software used by
energy-exploration companies to predict the location of oil and gas deposits.
Application: YASK, iso3dfd stencil
Geophysics
Image Source: US Dept. Energy
58

Seissol Seismic Solver*
SeisSol software simulates wave propagation and dynamic rupture based on the
arbitrary high-order accurate derivative discontinuous Galerkin method (ADER-
DG). Characteristics include tetrahedral meshes to approximate complex 3D model
geometries and rapid model generation use of elastic, viscoelastic and viscoplastic
material to approximate realistic geological subsurface properties. The code is
Open Source.
Application: SeisSol Seismic Solver.
Recipe: Download code and follow instructions
Geophysics
Images Source: Intel Labs
59

SPECFEM3D_GLOBE simulates the three-dimensional global and regional
seismic wave propagation based upon the spectral-element method (SEM).
It is a time-step algorithm which simulates the propagation of earth waves
given the initial conditions, mesh coordinates/ details of the earth crust.
Application: specfem3D_globe
Baseline Code: https://geodynamics.org/cig/software/specfem3d_globe/
Optimized Code: Pending check-in. Recipe: Runs out-of-the-box.
Specfem3d_globe*Geophysics
Image Source:
https://geodynamics.org/cig/soft
ware/specfem3d_globe/
60

ISO3DFD (3D acoustic Isotropic
finite difference)*
Iso3DFD - The Iso-3D 16th order Isotropic kernel is at the heart of RTM
algorithm. It plays a major role on accurate imaging of complex subsurfaces.
This kernel computes the wave propagation used in seismic imaging. The
code is in-house code.
Application: Iso3DFD (3D Acoustic Isotropic Finite Difference)
Code: Check for availability here
Recipe: -O3 -xMIC-AVX512 -fp-model fast –fma –qopenmp -lmemkind
x
y
z
Half Length
(HL) = 8
Energy Industry
Image Source: Intel
61

Sparse Matrix Vector Multiply*
Sparse matrix-vector multiplication (SpMV) is an important kernel for a
diverse set of applications in which systems with sparse pattern are used,
such as scientific computing, engineering, economic modeling, information
retrieval, oil & gas, weather consulting, animation, aero-spacing,
recommender systems in machine learning, and earthquake prediction.
Application: Sparse Matrix Vector Multiply
Code: Available here. Recipe: Check for availability here
Performance
62

High Performance Conjugate Gradients
The High Performance Conjugate Gradients (HPCG) Benchmark project is
an effort to create a new metric for ranking HPC systems. HPCG is intended
as a complement to the High Performance LINPACK (HPL) benchmark,
currently used to rank the TOP500 computing systems.
Application: High Performance Conjugate Gradients
Code: Available here. Recipe: Check for availability here
63
Performance
Image Source: Univ. Maryland, Baltimore Co.

Web Resources
• Public Intel® Xeon Phi™ Processor software proof points: https://software.intel.com/en-us/articles/intel-
xeon-phi-processor-applications-performance-proof-points
• Intel® Xeon Phi™ Processor Applications Showcase:
https://software.intel.com/sites/default/files/managed/b4/b0/Nov2016_Intel_Xeon_Phi_Processor_Showc
ase.pdf
• Intel® Xeon Phi™ Processor Performance Package:
http://competition.intel.com/Default.aspx?tabid=2273&DMXModule=6185&EntryId=14666&Command=C
ore_Download
• Intel® Xeon Phi™ Processor: http://www.intel.com/content/www/us/en/processors/xeon/xeon-phi-
detail.html
• Modern Code: https://software.intel.com/en-us/modern-code
• Intel® SSF Configurations: http://www.intel.com/ssfconfigurations
• Intel® Omni Path Architecture: http://www.intel.com/content/www/us/en/high-performance-computing-
fabrics/omni-path-architecture-fabric-overview.html
• Machine Learning: http://www.intel.com/content/www/us/en/analytics/machine-learning/overview.html
64

Intel® Xeon Phi™ processor (codenamed Knights Landing)applications Code and Recipe Guide

Intel® Xeon Phi™ processor (codenamed Knights Landing)applications Code and Recipe Guide

More Related Content

What's hot (9)

Similar to Intel® Xeon Phi™ processor (codenamed Knights Landing)applications Code and Recipe Guide (20)

Recently uploaded (20)