AccelerEyes has released version 1.5 of the Jacket GPU programming platform for MATLAB. Version 1.5 delivers an expanded Image Processing Library, enhancements to Jacket’s GFOR capability, performance enhancements across the platform, new capabilities for profiling applications, and the ability to compile full M-code functions to individual GPU kernels.
AccelerEyes develops Jacket, a software platform that delivers GPU computing power to desktop users of MATLAB and other very high-level languages. It enables faster prototyping and problem solving across a range of government, manufacturing, energy, media, biomedical, financial, and scientific research applications. The Jacket software platform enables accelerated performance for common arithmetic and linear algebra functionality across the complete line of CUDA-capable GPUs from NVIDIA.
New capabilities in Jacket 1.5 include two functions for profiling M-code and for compiling individual M-functions directly to the GPU:
- GPROFILE facilitates GPU and CPU profiling from the MATLAB console. GPROFILE records and reports on the time spent in MATLAB functions on the GPU and compares those results against comparable CPU timings. The results indicate sections of code that should or should not be run on the GPU as well as those that could benefit from vectorization or refactoring.
- GCOMPILE (and ARRAYFUN) enable Jacket programmers to define GPU kernels using M-code. Through this mechanism, Jacket is able to perform various additional optimizations, which it is otherwise unable to do in standard dynamic MATLAB mode.
Jacket automatically connects M-code to GPUs through its dynamic compiler technology allowing users to accelerate their applications across any CUDA-capable GPUs. Jacket-based MATLAB applications scale effortlessly across GeForce, Tesla C1060, or the new Tesla C2050 without modification. Jacket eliminates the need to manually port MATLAB code to C, C++, or CUDA to run on GPU-based workstations and clusters -- a productivity step that can take months to years to complete for computationally intensive problems. Jacket 1.5 is compatible across 32-bit and 64-bit versions of Linux, Windows, and Mac operating systems.
