ppt (Microsoft) - ICT

Download Report

Transcript ppt (Microsoft) - ICT

GPU Research Capabilities at Seneca

FSOSS

2012-10-26 Dr. Chris Szalwinski

Professor School of Information and Communication Technology Seneca College, Toronto, Canada

A Fresh Initiative

From Some Personal History To Heterogeneous Computing

2

The 80287

A Fresh Initiative

3

Floating-Point Co-Processor (1985) A Fresh Initiative

4

ATI 3D Rage II Co-Processor (1996) A Fresh Initiative

5

A Paradigm Shift In Programming

A Fresh Initiative

6

The Turn Towards Concurrency Paradigm Shift

7

Paradigm Shift

8

Can still increase

 transistor density – but it's getting more expensive

Paradigm Shift

9

Can still increase

 transistor density – but it's getting more expensive

Can't increase

 processor frequencies < 10 GHz chips

Paradigm Shift

10

Can still increase

 transistor density – but it's getting more expensive

Can't increase

  processor frequencies < 10 GHz chips power consumption – can't melt chips

Paradigm Shift

11

Paradigm Shift Can still increase

 transistor density – but it's getting more expensive

Can't increase

  processor frequencies < 10 GHz chips power consumption – can't melt chips

The Free Lunch is Over

 we can't just wait for improvement like we did before  we need new routes to improvement 12

Paradigm Shift

Use Different Computational Units For Distinctly Different Tasks

13

Heterogeneous Computing Intel Core i7 (2008), NVIDIA GeForce GTX580 (2010)

14

Heterogeneous Computing

15

Heterogeneous Computing

16

Serial processing

+ Heterogeneous Computing

Parallel processing 17

Heterogeneous Computing NVIDIA many-core GPUs vs Intel multi-core CPUs

 Floating point operations per sec (GFLOP/s)  Memory bandwidth (GB/s) 18

Industry Momentum STI (Sony + Toshiba + IBM)

 Broadband Cell Processor – CPU + GPU on one chip 19

Industry Momentum STI (Sony + Toshiba + IBM)

 Broadband Cell Processor – CPU + GPU on one chip

Intel

 Xeon Phi – MIC (Many Integrated Core) 20

Industry Momentum STI (Sony + Toshiba + IBM)

 Broadband Cell Processor – CPU + GPU on one chip

Intel

 Xeon Phi – MIC (Many Integrated Core)

AMD

 APUs (Fusion) – CPU + GPU on a single chip 21

Industry Momentum STI (Sony + Toshiba + IBM)

 Broadband Cell Processor – CPU + GPU on one chip

Intel

 Xeon Phi – MIC (Many Integrated Core)

AMD

  APUs (Fusion) – CPU + GPU on a single chip HSA Foundation (2012) – AMD + ARM + TI + Imagination + MediaTek + Samsung + Ateris + Multicore Ware + Apical + Sonics + Symbio + Vivante 22

Industry Momentum STI (Sony + Toshiba + IBM)

 Broadband Cell Processor – CPU + GPU on one chip

Intel

 Xeon Phi – MIC (Many Integrated Core)

AMD

  APUs (Fusion) – CPU + GPU on a single chip HSA Foundation (2012) – AMD + ARM + TI + Imagination + MediaTek + Samsung + Ateris + Multicore Ware + Apical + Sonics + Symbio + Vivante  Radeon – Discrete GPUs 23

Industry Momentum STI (Sony + Toshiba + IBM)

 Cell Processor – CPU + GPU on one chip

Intel

 Xeon Phi – MIC (Many Integrated Core)

AMD

  APUs (Fusion) – CPU + GPU on a single chip HSA Foundation (2012) – AMD + ARM + TI + Imagination + MediaTek + Samsung + Ateris + Multicore Ware + Apical + Sonics + Symbio + Vivante  Radeon – Discrete GPUs

NVIDIA – Discrete GPUs

 GeForce (digital gaming)   Quadro (engineering workstations - graphics) Tesla (scientific computations – double precision) 24

Industry Momentum Discrete GPUs - Add-in board shipments

25

Predictions

Industry Momentum

26

Industry Predictions Computer Graphics Market 1974-2015

27

Industry Predictions Computer Graphics Market 1974-2015

 Traditional processors + low-cost graphics processors enable combinations of science and entertainment 28

Industry Predictions Embedded Graphics Processors (EGPs) are killing off Integrated Graphics Processors (IGPs)

29

Industry Predictions Embedded Graphics Processors (EGPs) are no threat to Discrete Graphics

30

Programming Heterogeneous Computers Concurrency-Oriented Programming

Core Languages

   Fortran C C++ 31

Programming Heterogeneous Computers Concurrency-Oriented Programming (COP)

 

Core Languages

   Fortran C C++

Extensions for COP

   Cilk Plus (Intel) OpenCL (Khronos Group – AMD and HSA) CUDA  C/C++ (NVIDIA)  Fortran 2008, C-x86 (PGI)  DirectCompute (Microsoft) 32

Programming Heterogeneous Computers CUDA Teaching Centers in Ontario

McMaster University (2010)

 High Performance Parallel Computing on Graphical Processing Units – ECE709 – part of Master's Degree 

University of Toronto (2011)

 Special Topics in Software Engineering: Programming Massively Parallel Graphics Processors – ECE1724H – part of Master's Degree 

Seneca College (2012)

 Introduction to Parallel Programming – Professional Option – GPU610/DPS915 – CPA Diploma and BSD Degree 33

Programming Heterogeneous Computers

School of Information and Communications Technology (ICT) Our Capabilities and Plans

34

ICT Facilities Fully Equipped Teaching Classroom and Lab

 40 seats  38 CUDA enabled desktops with GTX480s (480 cores)

Maximus Workstation

 Quadro 600 for visualization  Tesla C2075 for computation

SCI-Net Research

 Accelerator Research Cluster – research testbed  8 x [2 Intel Xeon X5550 + 2 NVIDIA Tesla M2070] 35

The 80287

ICT Facilities

36

ICT Courses Introductory Course – Student Skill Set

 Solid tested background in both C and C++  Profile for computationally intensive code  Move critical code to the GPU using CUDA  Optimize to hide memory latency with computations

Programmer Training Workshops – on demand Advanced Course – (in the planning stage)

 Interactive Real-Time Computations + Visualization  Parallelizing Fortran Applications  OpenGL, DirectX Graphics Interoperability 37

Areas of Interest or Domain Expertise

    Big Data – Geocomputation Cognition – Cognitive Tutors Intrusion Detection – Information Security Finite Element Analysis – Soft Matter

ICT Faculty

38

Areas of Application (source: NVIDIA)

 Image Processing  Big Data Mining  Gaming  Advertising  Genetics  Quantum Chemistry  Mathematics  Product Design  Scientific Computing  Computational Finance

ICT Scope

39

GPU Research Capabilities at Seneca

FSOSS

2012-10-26 Dr. Chris Szalwinski

Professor School of Information and Communication Technology Seneca College, Toronto, Canada