Arm in HPCThe future of supercomputing starts now

Marcin Krzysztofik and Eric Lalardie

Arm

© 2018 Arm Limited

© 2018 Arm Limited

Arm’s business model (HPC focus)

Armv8.x and extensions,Neoverse IP roadmapSVE Scalable Vector Extension

The Cloud to Edge Infrastructure Foundation

for a World of 1T Intelligent Devices

© 2018 Arm Limited

© 2

01

8A

rm L

imit

ed

The embargo for this content presented at Arm Tech Day will lift on Wednesday, Feb 20th at 6 a.m. Pacific Time. Corresponding UK time is: Wednesday, Feb 20th 2 p.m. BST

Each generation brings faster performance and new infrastructure specific features

16nm

CosmosPlatform

7nm

NeoverseN1

Platform

7nm+

ZeusPlatform

PoseidonPlatform

5nm

20212020

2019Today

30% Faster System Performance per Generation + New Features

World-class Neoverse Ecosystem

© 2018 Arm Limited

A M P E R E

Tencent Cloud orange

© 2018 Arm Limited

Vanguard Astra by HPE (Top 500 system)• 2,592 HPE Apollo 70 compute nodes

• 5,184 CPUs, 145,152 cores, 2.3 PFLOPs (peak)

• Cavium Thunder-X2 ARM SoC, 28 core, 2.0 GHz

• Memory per node: 128 GB (16 x 8 GB DR DIMMs)

• Aggregate capacity: 332 TB, 885 TB/s (peak)

• Mellanox IB EDR, ConnectX-5 • 112 36-port edges, 3 648-port spine

switches

• Red Hat RHEL for Arm

• HPE Apollo 4520 All–flash Lustre storage• Storage Capacity: 403 TB (usable)

• Storage Bandwidth: 244 GB/s

© 2018 Arm Limited

Arm HPC Software EcosystemC

luste

r Man

agem

en

t Too

ls:B

right, H

PE C

MU

, xCat, W

arewu

lf

Silicon Suppliers:Marvell, Fujitsu, Huawei

Mellanox Linux OS Distro of choice:RHEL, SUSE, CENTOS,…

Arm Server Ready Platform:Standard OS compatible FW and RAS features

HPC Applications: Open-source, Owned, and Commercial ISV codes

Job schedulers and Resource Management:

SLURM, IBM LSF, Altair PBS Pro, etc.

Programming Languages:

Fortran, C, C++via

GNU, LLVM, Arm & OEMs

Debug and performance

analysis tools:Arm Forge,

Rogue Wave, TAU, etc.

Filesystems: BeeGFS,

LUSTRE, ZFS, HDFS, GPFS

App/ISA specific optimizations, optimized libs and intrinsics:Arm PL, BLAS, FFTW, etc.

OEM/ODM’s:Cray, HPE, ATOS-Bull, Fujitsu, Gigabyte, Inventec, Foxconn

Communication Stacks and run-times:Mellanox IB/OFED/HPC-X, OpenMPI, MPICH, MVAPICH2, OpenSHMEM, OpenUCX, HPE MPI

Parallelism standards:OpenMP

(omp / gomp), MPI, SHMEM(see below)

User-space utilities, scripting,

containers, and other packages:

Singularity,Openstack,

OpenHPC, Python, NumPy, SciPy, etc.

© 2018 Arm Limited

Our solution for any architecture, at any scaleThe world-leading solution for better, faster and more efficient applications

Arm Performance Libraries

BLAS, LAPACK and FFT

Arm Compiler for HPC

Linux user space compiler for HPC applications

Arm Performance Reports

Interoperable application performance insight

Arm Allinea Studio All-inclusive development toolkit for Arm hardware

Arm Forge Professional

Multi-node interoperable profiler and debugger

Arm MAP

Speed-up applications with a lightweight scalable profiler

Arm DDT

Slash your time to debug on any hardware, at any scale.

Arm Performance Reports

Find the most efficient settings for your workloads.

Cross-Platforms Tools Debug, optimise and analyse on any architecture

Arm Forge Professional

Arm DDT and MAP in One Single Package

© 2018 Arm Limited

Key highlights in Forge & Performance ReportsVersion 19.0, December 2018

Forge Performance ReportsDDT MAP

PackagingCreation of Arm Allinea Studio

A new solution for aarch64 platforms that includes the Arm Compiler, Arm Performance Libraries, and the former Allinea tools!

PlatformsFull support for IBM systemsArm v8 supportCUDA 9 support

Full support for IBM systemsArm v8 supportCUDA 9 support

Full support for IBM systemsArm v8 supportCUDA 9 support

ImprovementsUsability ImprovementsMemory debugging optimizations

Optimizations for many-core systems

Optimizations for many-core systems

New FeaturesCombined C/C++/Fortran and Python Debugging

Python profilingBackfill Custom MetricsOn-kernel GPU profilingAbility to profile selected ranks

Python performance analysisAbility to profile selected ranks

© 2018 Arm Limited

Arm HPC Community

Communication Portals• Arm resources: www.arm.com

• HPC Ecosystem https://developer.arm.com/hpc

• HPC Blogs, Forum: www.community.arm.com/tools/HPC

Arm HPC User Group Community

• Gitlab HPC Packages Wiki (software ecosystem)

• Arm-HPC @ Groups.IO (<=NEW)

Supporting Arm HPC Community end-users and developers.

Thank you

© 2018 Arm Limited

Marcin Krzysztofik & Eric LalardieArm