University GPU Club Tues 29 Oct

Slides:



Advertisements
Similar presentations
Founded in 2010: UCL, Southampton, Oxford and Bristol Key Objectives of the Consortium: Prove the concept of shared, regional e-infrastructure services.
Advertisements

Prasanna Pandit R. Govindarajan
Instructor Notes This lecture describes the different ways to work with multiple devices in OpenCL (i.e., within a single context and using multiple contexts),
EPrints: Sustainability Panel Les Carr. Mission Alignment - Context Intelligence, Agents, Multimedia Group, School of Electronics and Computer Science,
Research Computing and Facilitating Services CLMS Symposium 28 th June 2012 Clare Gryce Head of Research Computing & Facilitating Services.
Vectors, SIMD Extensions and GPUs COMP 4611 Tutorial 11 Nov. 26,
The Development of Mellanox - NVIDIA GPUDirect over InfiniBand A New Model for GPU to GPU Communications Gilad Shainer.
1 Computational models of the physical world Cortical bone Trabecular bone.
Monte-Carlo method and Parallel computing  An introduction to GPU programming Mr. Fang-An Kuo, Dr. Matthew R. Smith NCHC Applied Scientific Computing.
APARAPI Java™ platform’s ‘Write Once Run Anywhere’ ® now includes the GPU Gary Frost AMD PMTS Java Runtime Team.
Computing with Accelerators: Overview ITS Research Computing Mark Reed.
What is GPGPU? Many of these slides are taken from Henry Neeman’s presentation at the University of Oklahoma.
GPU System Architecture Alan Gray EPCC The University of Edinburgh.
XEON PHI. TOPICS What are multicore processors? Intel MIC architecture Xeon Phi Programming for Xeon Phi Performance Applications.
Early Linpack Performance Benchmarking on IPE Mole-8.5 Fermi GPU Cluster Xianyi Zhang 1),2) and Yunquan Zhang 1),3) 1) Laboratory of Parallel Software.
Hybrid Redux: CUDA / MPI 1. CUDA / MPI Hybrid – Why? 2  Harness more hardware  16 CUDA GPUs > 1!  You have a legacy MPI code that you’d like to accelerate.
GPGPU Introduction Alan Gray EPCC The University of Edinburgh.
HPCC Mid-Morning Break High Performance Computing on a GPU cluster Dirk Colbry, Ph.D. Research Specialist Institute for Cyber Enabled Discovery.
PVOCL: Power-Aware Dynamic Placement and Migration in Virtualized GPU Environments Palden Lama, Xiaobo Zhou, University of Colorado at Colorado Springs.
FSOSS Dr. Chris Szalwinski Professor School of Information and Communication Technology Seneca College, Toronto, Canada GPU Research Capabilities.
LinkSCEEM-2: A computational resource for the development of Computational Sciences in the Eastern Mediterranean Mostafa Zoubi SESAME SESAME – LinkSCEEM.
Evaluating GPU Passthrough in Xen for High Performance Cloud Computing Andrew J. Younge 1, John Paul Walters 2, Stephen P. Crago 2, and Geoffrey C. Fox.
Programming with CUDA, WS09 Waqar Saleem, Jens Müller Programming with CUDA and Parallel Algorithms Waqar Saleem Jens Müller.
Weekly Report Start learning GPU Ph.D. Student: Leo Lee date: Sep. 18, 2009.
What’s New in the Cambridge High Performance Computer Service? Mike Payne Cavendish Laboratory Director - Dr. Paul Calleja.
CS 732: Advance Machine Learning Usman Roshan Department of Computer Science NJIT.
Panda: MapReduce Framework on GPU’s and CPU’s
HPCC Mid-Morning Break Dirk Colbry, Ph.D. Research Specialist Institute for Cyber Enabled Discovery Introduction to the new GPU (GFX) cluster.
ORIGINAL AUTHOR JAMES REINDERS, INTEL PRESENTED BY ADITYA AMBARDEKAR Overview for Intel Xeon Processors and Intel Xeon Phi coprocessors.
Introduction to LinkSCEEM and SESAME 15 June 2014, ibis Hotel, Amman - Jordan Presented by Salman Matalgah Computing Group leader SESAME.
GPU Programming with CUDA – Accelerated Architectures Mike Griffiths
High Performance Computing G Burton – ICG – Oct12 – v1.1 1.
DRAFT 1 Institutional Research Computing at WSU: A community-based approach Governance model, access policy, and acquisition strategy for consideration.
David Luebke NVIDIA Research GPU Computing: The Democratization of Parallel Computing.
GPU Performance Prediction GreenLight Education & Outreach Summer Workshop UCSD. La Jolla, California. July 1 – 2, Javier Delgado Gabriel Gazolla.
By Arun Bhandari Course: HPC Date: 01/28/12. GPU (Graphics Processing Unit) High performance many core processors Only used to accelerate certain parts.
GPU Computing April GPU Outpacing CPU in Raw Processing GPU NVIDIA GTX cores 1.04 TFLOPS CPU GPU CUDA Architecture Introduced DP HW Introduced.
Taking the Complexity out of Cluster Computing Vendor Update HPC User Forum Arend Dittmer Director Product Management HPC April,
Use of GPUs in ALICE (and elsewhere) Thorsten Kollegger TDOC-PG | CERN |
OpenCL Sathish Vadhiyar Sources: OpenCL overview from AMD OpenCL learning kit from AMD.
S&T IT Research Support 11 March, 2011 ITCC. Fast Facts Team of 4 positions 3 positions filled Focus on technical support of researchers Not “IT” for.
Introducing collaboration members – Korea University (KU) ALICE TPC online tracking algorithm on a GPU Computing Platforms – GPU Computing Platforms Joohyung.
Linchuan Chen. 图形处理器( Graphics Processing Unit ), 是一种专门用来处理在个人电脑、工作站或游 戏机上图像运算工作的微处理器。 图形处理器使显卡减少了对中央处理器的依赖, 并分担了部分原本是由中央处理器所担当的工 作 Efficient at manipulating.
Experts in numerical algorithms and HPC services Compiler Requirements and Directions Rob Meyer September 10, 2009.
Carlo del Mundo Department of Electrical and Computer Engineering Ubiquitous Parallelism Are You Equipped To Code For Multi- and Many- Core Platforms?
GPUs: Overview of Architecture and Programming Options Lee Barford firstname dot lastname at gmail dot com.
Compiler and Runtime Support for Enabling Generalized Reduction Computations on Heterogeneous Parallel Configurations Vignesh Ravi, Wenjing Ma, David Chiu.
ITCS 4/5145 Parallel Programming, UNC-Charlotte, B. Wilkinson, Dec 26, 2012outline.1 ITCS 4145/5145 Parallel Programming Spring 2013 Barry Wilkinson Department.
Biryaltsev E.V., Galimov M.R., Demidov D.E., Elizarov A.M. HPC CLUSTER DEVELOPMENT AND OPERATION EXPERIENCE FOR SOLVING THE INVERSE PROBLEMS OF SEISMIC.
Desktop Introduction. MASSIVE is … A national facility $8M of investment over 3 years Two high performance computing facilities, located at the Australian.
Software Sustainability Institute Building sustainable software for science … why good code is only the beginning 10 April 2013, EGI.
The National Grid Service Mike Mineter.
GPGPU introduction. Why is GPU in the picture Seeking exa-scale computing platform Minimize power per operation. – Power is directly correlated to the.
How to use HybriLIT Matveev M. A., Zuev M.I. Heterogeneous Computations team HybriLIT Laboratory of Information Technologies (LIT), Joint Institute for.
Parallel Computers Today Oak Ridge / Cray Jaguar > 1.75 PFLOPS Two Nvidia 8800 GPUs > 1 TFLOPS Intel 80- core chip > 1 TFLOPS  TFLOPS = floating.
Early Experience with Applications on POWER8 Mike Ashworth, Rob Allan, Rupert Ford, Xiaohu Guo, Mark Mawson, Jianping Meng, Andrew Porter STFC Hartree.
Hartree Centre systems overview. Public nameInternal nameTechnologyService type Blue WonderInvictax86 SandyBridgeproduction Blue WonderNapierx86 IvyBridgeproduction.
Introduction to Data Analysis with R on HPC Texas Advanced Computing Center Feb
Earth System Modelling: an HPC perspective Mike Ashworth & Rupert Ford Scientific Computing Department and STFC Hartree Centre STFC Daresbury Laboratory.
11 Brian Van Straalen Portable Performance Discussion August 7, FASTMath SciDAC Institute.
Sobolev(+Node 6, 7) Showcase +K20m GPU Accelerator.
NIIF HPC services for research and education
HPC Roadshow Overview of HPC systems and software available within the LinkSCEEM project.
GPU Computing Jan Just Keijser Nikhef Jamboree, Utrecht
LinkSCEEM-2: A computational resource for the development of Computational Sciences in the Eastern Mediterranean Mostafa Zoubi SESAME Outreach SESAME,
UK Grid: Moving from Research to Production
Overview of HPC systems and software available within
The Free Lunch Ended 7 Years Ago
CUBAN ICT NETWORK UNIVERSITY COOPERATION (VLIRED
CSE 502: Computer Architecture
Presentation transcript:

University GPU Club Tues 29 Oct

Today’s Agenda Introduction GPU Computing & The Future of HPC Tim Lanfear, NVIDIA Minke Whale Turbine Impact Using CUDA Driven Smoothed Particle Hydrodynamics Stephen Longshaw, MACE Refreshments

GPU News

NVIDIA CUDA Research Centre – Access to NVIDIA expertise – A K20 card – Discounts on equipment – Seeding programme for new GPUs – Bespoke live online training – “Intermediate CUDA”

Why GPU?

GPU General Purpose Graphics Processing Unit – Games: faster & faster graphics cards – Not just for viz

GPU General Purpose Graphics Processing Unit – Games: faster & faster graphics cards – Not just for viz – Decent (TFLOP) performance for compute – Certain types of compute

GPU General Purpose Graphics Processing Unit – Games: faster & faster graphics cards – Not just for viz – Decent (TFLOP) performance for compute – Certain types of compute GPU Programming – GPU is a discrete card (PCI-e) – OpenCL, CUDA (& extensions)

High Performance? developer.amd.com John D. Owens, UC Davis NB – this info slide from

GPU Resources (inc. other emerging tech…)

Local GPU Resources CSF contributors – NVIDIA K20s (MACE) – 2x NVIDIA M2070 – 5x NVIDIA M GPU per Intel node – 16x NVIDIA M2050s (MACE contribution) 2 GPUs per AMD node InfiniBand interconnect Redqueen – 3x AMD FirePro V7800

Local GPU Resources CSF contributors – NVIDIA K20s (MACE) – 2x NVIDIA M2070 – 5x NVIDIA M GPU per Intel node – 16x NVIDIA M2050s (MACE contribution) 2 GPUs per AMD node InfiniBand interconnect

National GPU Resources EMERALD (e-Infrastructure South) – 372x NVIDIA M2090s (3 or 8 GPUs per node) – Pilot users: Bristol, Oxford, Southampton, UCL – Other UK research institutions can enquire

National GPU Resources EMERALD (e-Infrastructure South) – 372x NVIDIA M2090s (3 or 8 GPUs per node) – Pilot users: Bristol, Oxford, Southampton, UCL – Other UK research institutions can enquire Daresbury (to be confirmed if still available) – 8x Tesla S1070 nodes 4x GPUs per node – 2x AMD nodes 3x AMD FireStream 9270s per node

Emerging Tech… Intel Xeon Phi system “xenomorph” – 2x 7120P (top end, new range) – Code from Xeon runs on Xphi – Can use FORTRAN, C, MPI, OpenMP Emerging Tech Conference 2015 – International speakers – Overcoming barriers to performance

GPU Training resources

University of Manchester IT Services Training Programme – Intermediate CUDA (NVIDIA, hosted by UoM) Weds 30 October – Introductory OpenCL Tues 12 Nov Basics / Host in-depth / Queues Intermediate course in Feb/Mar

NAG/HECToR training Training provided for HECToR Service – Dates TBC An Introduction to CUDA Programming An Introduction to OpenCL Programming – Free to UK academics if work under remit of: EPSRC, NERC and BBSRC –

Conferences

University GPU Club Making the most of GPUs (and other tech) – How to perform meaningful comparisons (across h/w) – Share lessons learned – Share equipment? Discuss emerging tech Community driven – 300 researchers, regular meetings, list – Training inc. from vendors

University GPU Club Tues 29 OctNVIDIA & MACE Weds 30 OctIntermediate CUDA training by NVIDIA Tues 12 NovIntro to OpenCL Weds 13 Nov (with CAS)Atmospheric Modelling with GPUs and FPGAs (John Michalakes/NOAA & Maxeler)