GPU Lynx: a Dynamic Instrumentation System for Data-Parallel Applications on GPGPU Architectures

Posted by on Sep 21, 2012 in Open Source, Software release

GPU Lynx is a dynamic instrumentation framework intended to facilitate the creation and insertion of PTX instrumentation procedures in CUDA applications. This project was developed by Naila Farooqui and is based on concepts described in her paper presented at ISPASS 2012. GPU Lynx utilizes GPU Ocelot’s PTX internal representation and provides a custom interposer layer for the CUDA Runtime API as well as a compilation path from C to PTX. Visit the project site for GPU Lynx:...

read more

Ocelot and SST-MacSim Tutorial at ISCA 2012

Posted by on Apr 18, 2012 in Conference, News, Tutorial

We are pleased to announce that we will be presenting a joint GPU Ocelot and SST-MacSim tutorial during ISCA 2012 on June 10, 2012, in Portland. An outline of the GPU Ocelot part of the tutorial is as follows: Overview of Ocelot Ocelot: Architecture CUDA Runtime API implementation Ocelot Device Abstraction Device interface Memory allocations Parallel Thread Execution (PTX) IR Introduction to ISA and internal representation Writing PTX transformations and Pass Manager Details of the parser and emitter Ocelot: Supported Devices PTX...

read more

GPU Ocelot in the News

Posted by on Apr 13, 2012 in News

NVIDIA Names Georgia Institute of Technology a CUDA Center of Excellence NVIDIA Opens Cuda Compiler Source Code NVIDIA Opens Up CUDA Compiler NVIDIA CUDA Compiler Source Code Available For Download Haicheng Wu awarded NVIDIA Fellowshiop NVIDIA Graduate Fellowship Awards...

read more

GPU Ocelot tutorial at Microsoft Research

Posted by on Mar 15, 2012 in News, Tutorial

GPU Ocelot tutorial at Microsoft Research on March 16, 2012. This tutorial will describe Ocelot’s implementation of the CUDA Runtime API, its abstract device interface, PTX internal representation and compiler analysis data structures. It will feature demos executing Ocelot programs, applying PTX analyses and transformations, and extracting application characteristics. Presentation Slides Website Video of tutorial Thanks to Aaron Smith and Amir Hormati for arranging this...

read more

Joint GPU Ocelot and MACSIM Tutorial at HPCA

Posted by on Feb 3, 2012 in Tutorial

MacSim is a heterogeneous architecture simulator developed by Professor Hysoon Kim and her group at Georgia Tech. The Structural Simulation Toolkit is a parallel simulation infrastructure developed at Sandia National Laboratories (Albuquerque) led by Dr. Arun Rodrigues. SST-MacSim represents MacSim components that execute using the SST infrastructure. Ocelot is used to generate traces for execution by SST-MacSim architecture models. We presented a joint GPU Ocelot, SST, and MACSIM Tutorial at HPCA 2012. This tutorial describes how GPU Ocelot...

read more

Need Your Help for Windows Support

Posted by on Feb 2, 2012 in Open Source

The Ocelot source tree currently supports building on windows, however, we do not have development machines that can be used for building and testing on windows.  We need volunteers with access to VS2010 who would be willing to checkout the codebase build it on windows, and create an installer for the ocelot libraries. If you are interested, please respond to this thread:

read more

GPU Ocelot

Posted by on Jan 11, 2012 in News

Welcome to the project site of the GPU Ocelot project. We are in the process of adding content related to our research in GPU computing. Meanwhile, visit the GPU Ocelot Google Code site for more information. GPU Ocelot source documentation available at:

read more