Introduc)on to High Performance Compu)ng Advanced Research Computing

Transcription

1 Introduc)on to High Performance Compu)ng Advanced Research Computing

2 Outline What cons)tutes high performance compu)ng (HPC)? When to consider HPC resources What kind of problems are typically solved? What are the components of HPC? What resources are available? Overview of HPC Resources at Virginia Tech 2

3 Should I Pursue HPC for my Problem? Are local resources insufficient to meet your needs? Very large jobs Very many jobs Large data Do you have na)onal collaborators? Share projects between different en))es Convenient mechanisms for data sharing 3

4 Who Uses HPC? Training (51) 2% Earth Sci (29) 2% ScienEfic CompuEng (60) 2% Chemistry (161) 7% Chemical, Thermal Sys (89) 8% Materials Research (131) 9% Atmospheric Sciences (72) 11% Physics (91) 19% Molecular Biosciences (271) 17% Astronomical Sciences (115) 13% >2 billion cpu- hours allocated 1400 alloca)ons 350 ins)tu)ons 32 research domains

5 Learning Curve Linux: Command- line interface Scheduler: Shares resources among mul)ple users Parallel Compu)ng: Need to parallelize code to take advantage of supercomputer s resources Third party programs or libraries make this easier

6 Popular SoYware Packages Molecular Dynamics: Gromacs, LAMMPS CFD: OpenFOAM, Ansys Finite Elements: Deal II, Abaqus Chemistry: VASP, Gaussian Climate: CESM Bioinforma)cs: Mothur, QIIME, MPIBLAST Numerical Compu)ng/Sta)s)cs: R, Matlab Visualiza)on: ParaView, Ensight

7 WHAT IS PARALLEL COMPUTING? 8

8 Parallel Compu)ng 101 Parallel compu)ng: use of mul)ple processors or computers working together on a common task. Each processor works on its sec)on of the problem Processors can exchange informa)on Grid of Problem to be solved CPU #1 works on this area of the problem CPU #2 works on this area of the problem y exchange exchange exchange CPU #3 works on this area of the problem exchange CPU #4 works on this area of the problem x 9

9 Why Do Parallel Compu)ng? Limits of single CPU compu)ng performance available memory I/O rates Parallel compu)ng allows one to: solve problems that don t fit on a single CPU solve problems that can t be solved in a reasonable )me We can solve larger problems faster more cases 10

10 A Change in Moore s Law

11 Parallelism is the New Moore s Law Power and energy efficiency impose a key constraint on design of micro- architectures Clock speeds have plateaued Hardware parallelism is increasing rapidly to make up the difference

12 WHAT DOES A MODERN SUPERCOMPUTER LOOK LIKE? 13

13 Essential Components of HPC Supercompu)ng resources Storage Visualiza)on Data management Network infrastructure Support 15

14 Blade : Rack : System 1 node : 2 x 8 cores = 16 cores 1 chassis : 10 nodes = 160 cores 1 rack (frame) : 4 chassis = 640 cores system : 10 racks = 6,400 cores x 10 x 4

15 Shared and distributed memory Memory M M M M M P P P P P P P P P Network P All processors have access to a pool of shared memory Access )mes vary from CPU to CPU in NUMA systems Example: SGI UV, CPUs on same node Memory is local to each processor Data exchange by message passing over a network Example: Clusters with single- socket blades 18

16 HPC Trends Memory Memory M P GPU Architecture Single core Mul)core GPU Cluster Code Serial OpenMP, Pthreads CUDA, OpenACC MPI

17 How are accelerators different? Intel Xeon E (CPU) Intel Xeon Phi 5110P (MIC) Nvidia Tesla K20X (GPU) Cores SMX Logical Cores ,688 CUDA cores Frequency 2.60 GHz 1.05 GHz 0.74 MHz GFLOPs (double) 333 1,010 1,317 Memory 64 GB 8GB 6GB Memory B/W 51.2GB/s 320GB/s 250GB/s

18 Mul)- core systems Memory Memory Memory Memory Memory Network Current processors place mul)ple processor cores on a die Communica)on details are increasingly complex Cache access Main memory access Quick Path / Hyper Transport socket connec)ons Node to node connec)on via network

19 Accelerator- based Systems Memory Memory Memory Memory G P U G P U G P U G P U Network Calcula)ons made in both CPUs and Graphical Processing Unit No longer limited to single precision calcula)ons Load balancing cri)cal for performance Requires specific libraries and compilers (CUDA, OpenCL) Co- processor from Intel: MIC (Many Integrated Core)

20 Batch Submission Process Login Node Compute Nodes Internet ssh qsub job Queue Master Node C1 C2 C3 mpirun np #./a.out Queue: Job script waits for resources. Master: Compute node that executes the job script, launches all MPI processes. ibrun./a.out

21 ARC OVERVIEW 24

22 (ARC) Unit within the Office of the Vice President of Informa)on Technology Provide centralized resources for: Research compu)ng Visualiza)on Staff to assist users Website: hmp://

23 Goals Advance the use of compu)ng and visualiza)on in VT research Centralize resource acquisi)on, maintenance, and support for research community Provide support to facilitate usage of resources and minimize barriers to entry Enable and par)cipate in research collabora)ons between departments

24 Personnel Associate VP for Research Compu)ng: Terry Herdman Director, HPC: Vijay Agarwala Director, Visualiza)on: Nicholas Polys Computa)onal Scien)sts Jus)n Krome)s James McClure Brian Marshall Srinivas Yarlanki Srijith Rajamohan

25 Personnel (Con)nued) System Administrators Tim Rhodes Chris Snapp Brandon Sawyers Vis & Virtual Reality Specialist: Wole Oyekoya Business Manager: Alana Romanella User Support GRAs: Umar Kalim and Di Zhang

26 Computa)onal Resources Name BlueRidge HokieSpeed HokieOne Ithaca Key Features, Uses Large- scale CPU or MIC GPU Shared Memory Beginners, MATLAB Available March 2013 Sept 2012 Apr 2012 Fall 2009 Theore)cal Peak (TFlops/s) Nodes N/A 79 Cores 6,528 2, Cores/Node N/A* 8 Accelerators/ Coprocessors 260 Intel Xeon Phi 8 Nvidia K40 GPU 408 Nvidia Tesla GPU N/A N/A Memory Size 27.3 TB 5.0 TB 2.62 TB 2 TB Memory/Core 4 GB* 2 GB 5.3 GB 3 GB* Memory/Node 64 GB* 24 GB N/A* 24 GB*

27 Visualiza)on Resources VisCube: 3D immersion environment with three 10ʹ by 10ʹ walls and a floor of stereo projec)on screens DeepSix: Six )led monitors with combined resolu)on of ROVR Stereo Wall AISB Stereo Wall

28 Gewng Started on ARC Systems 1. Review ARC s system specifica)ons and choose the right system(s) for you a. Specialty soyware 2. Apply for an account online the Advanced Research Compu)ng website 3. When your account is ready, you will receive confirma)on from ARC s system administrators

29 Resources ARC Website: hmp:// ARC Compute Resources & Documenta)on: hmp:// New Users Guide: hmp:// Frequently Asked Ques)ons: hmp:// Linux Introduc)on: hmp://

30 Thank you. Ques)ons?