From the October 2014 Issue
GPU-Aware MPI on RDMA-Enabled Clusters: Design, Implementation and Evaluation
By Hao Wang, Sreeram Potluri, Devendar Bureddy, Carlos Rosales, and Dhabaleswar K. (DK) Panda
Designing high-performance and scalable applications on GPU clusters requires tackling several challenges. The key challenge is the separate host memory and device memory, which requires programmers to use multiple programming models, such as CUDA and MPI, to operate on data in different memory spaces. This challenge becomes more difficult to tackle when non-contiguous data in multidimensional structures is used by real-world applications. These challenges limit the programming productivity and the application performance. We propose the GPU-Aware MPI to support data communication from GPU to GPU using standard MPI. It unifies the separate memory spaces, and avoids explicit CPU-GPU data movement and CPU/GPU buffer management. It supports all MPI datatypes on device memory with two algorithms: a GPU datatype vectorization algorithm and a vector based GPU kernel data pack and unpack algorithm. A pipeline is designed to overlap the non-contiguous data packing and unpacking on GPUs, the data movement on the PCIe, and the RDMA data transfer on the network. We incorporate our design with the open-source MPI library MVAPICH2 and optimize a production application: the multiphase 3D LBM. Besides the increase of programming productivity, we observe up to 19.9 percent improvement in application-level performance on 64 GPUs of the Oakley supercomputer.
Editorials and Announcements
- According to Thomson Reuters' 2013 Journal Citation Report, TPDS has an impact factor of 2.173.
- TPDS celebrates its 25th Anniversary. Editor-in-Chief David A. Bader says, "Congratulations to TPDS on its Silver Jubilee! For 25 years, TPDS has been the parallel and distributed computing community's flagship journal for research breakthroughs!"
- Get Your Journals as eBooks for Free
- Print on Demand is Now Available for OnlinePlus Titles
- eBooks of issues of TPDS can now be downloaded from the Computer Society Digital Library
- State of the Journal (Jan 2014)
- Editor's Note: EIC Farewell and New EIC Introduction (Dec 2013)
- Editor's Note (Nov 2013)
- Editor's Note (Jan 2013)
- Editor's Note (April 2012)
- Editor's Note (January 2012)
- Editorial: Media Center (November 2011)
- Editor's Note: How to Write Research Articles in Computing and Engineering Disciplines by Ivan Stojmenovic
- Full Supplemental PDF of Editor's Note: How to Write Research Articles in Computing and Engineering Disciplines by Ivan Stojmenovic and Veljko Milutinovic (PDF)
- Special Issue on Trust, Security, and Privacy in Parallel and Distributed Systems (Feb 2014)
- Special Issue on Cloud Computing (June 2013)
- Special Issue on Cyber-Physical Systems (CPS) (Sept 2012)
- Special Section on Many-Task Computing (June 2011)
Access recently published TPDS articles
Subscribe to the RSS feed of latest TPDS content added to the digital library.
Sign up for the Transactions Connection newsletter.
Listen to the OnlinePlus Podcast: Computer Society Publishing—two more titles migrate to OnlinePlus™ in 2012.
In this podcast, VP of Publications, David Alan Grier talks about Transactions on Mobile Computing and Transactions on Parallel and Distributed Systems migrating to OnlinePlus™.
TPDS is indexed in ISI
IEEE Transactions on Parallel and Distributed Systems (TPDS) is a scholarly archival journal published monthly. Parallelism and distributed computing are foundational research and technology to rapidly advance computer systems and their applications.
Read the full scope of TPDS