400 likes | 418 Views
GPU Research Capabilities at Seneca. A Fresh Initiative. From Some Personal History To Heterogeneous Computing. A Fresh Initiative. The 80287. A Fresh Initiative. Floating-Point Co-Processor (1985). A Fresh Initiative. ATI 3D Rage II Co-Processor (1996). A Fresh Initiative.
E N D
A Fresh Initiative • From • Some Personal History • To • Heterogeneous Computing
A Fresh Initiative • The 80287
A Fresh Initiative • Floating-Point Co-Processor (1985)
A Fresh Initiative • ATI 3D Rage II Co-Processor (1996)
A Fresh Initiative • A Paradigm Shift • In Programming
Paradigm Shift • The Turn Towards Concurrency
Paradigm Shift • Can still increase • transistor density – but it's getting more expensive
Paradigm Shift • Can still increase • transistor density – but it's getting more expensive • Can't increase • processor frequencies < 10 GHz chips
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
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
Paradigm Shift • Use Different • Computational Units • For • Distinctly Different Tasks
Heterogeneous Computing • Intel Core i7 (2008), NVIDIA GeForce GTX580 (2010)
Heterogeneous Computing Parallel processing + • Serial processing
Heterogeneous Computing • NVIDIA many-core GPUs vs Intel multi-core CPUs • Floating point operations per sec (GFLOP/s) • Memory bandwidth (GB/s)
Industry Momentum • STI (Sony + Toshiba + IBM) • Broadband Cell Processor – CPU + GPU on one chip
Industry Momentum • STI (Sony + Toshiba + IBM) • Broadband Cell Processor – CPU + GPU on one chip • Intel • Xeon Phi – MIC (Many Integrated Core)
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
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
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
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)
Industry Momentum • Discrete GPUs - Add-in board shipments
Predictions Industry Momentum
Industry Predictions • Computer Graphics Market 1974-2015
Industry Predictions • Computer Graphics Market 1974-2015 • Traditional processors + low-cost graphics processors enable combinations of science and entertainment
Industry Predictions • Embedded Graphics Processors (EGPs) are killing off Integrated Graphics Processors (IGPs)
Industry Predictions • Embedded Graphics Processors (EGPs) are no threat to Discrete Graphics
Programming Heterogeneous Computers • Concurrency-Oriented Programming • Core Languages • Fortran • C • C++
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)
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
School of Information and Communications Technology (ICT) Our Capabilities and Plans Programming Heterogeneous Computers
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]
ICT Facilities The 80287
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
ICT Faculty • Areas of Interest or Domain Expertise • Big Data – Geocomputation • Cognition – Cognitive Tutors • Intrusion Detection – Information Security • Finite Element Analysis – Soft Matter
ICT Scope • Areas of Application (source: NVIDIA) • Image Processing • Big Data Mining • Gaming • Advertising • Genetics • Quantum Chemistry • Mathematics • Product Design • Scientific Computing • Computational Finance