Tutorial I:   (Half-day, morning of November 6)    Heterogeneous Computing with GPU Ocelot (by Georgia Tech )

Tutorial II:  (Full Day, November 6)  Energy Efficient Datacenters and Systems (by IBM Research - Austin)

Tutorial Descriptions

Tutorial I: Heterogeneous Computing with GPU Ocelot

  GPU Ocelot is an open-source dynamic JIT compilation framework for GPU compute applications targeting a > range of GPU and non-GPU execution targets. Ocelot supports CUDA applications and provides an implementation of the CUDA Runtime API enabling seamless integration with existing CUDA applications. Its JIT compiler supports four backend execution targets - (1) an emulator that implements NVIDIAs Parallel Thread Execution (PTX) instruction set architecture, (2) NVIDIA, (3) AMD GPUs, and (4) a translator to LLVM for efficient execution of GPU kernels on multicore CPUs. A back-end targeting vector cores is in development. Existing CUDA applications are seamlessly supported. Ocelot facilitates research and development on several fronts. First, Ocelot improves developer productivity of GPU compute applications by providing an infrastructure for building event trace analyzers using the emulator. This tutorial will describe how analyzers can be built and cover several existing analyzers for (1) correctness checks, (2) debugging and (3) performance tuning. Second, as a JIT compiler infrastructure, Ocelot provides facilities for compiler research including interfaces to an internal representation of PTX programs in support of optimization passes for massively data parallel computer kernels. Third, with an open source re-implementation of the CUDA runtime, Ocelot enables research into scheduling, resource allocation, and operating systems. Finally, Ocelot enables research in heterogeneous architectures via trace generation interfaces, PTX emulation and support for detailed workload characterization on GPU and CPU devices.

Tutorial II: Energy Efficient Datacenters and Systems