On Synchronisation and Memory Reuse in OpenSHMEM, In: OpenSHMEM 2016: OpenSHMEM and Related Technologies. Enhancing OpenSHMEM for Hybrid Environments
- Extreme Scale Systems Center, Oak Ridge National Laboratory, Oak Ridge, TN (United States)
OpenSHMEM is an open standard for PGAS libraries that provides one-sided communication semantics. Since the standardisation process was completed in 2012, the OpenSHMEM API has seen a rapid succession of proposed extensions. Among these extensions is the addition of teams of processing element (PEs) for greater flexibility in defining PE subsets for problem decomposition. Adding further to this, spaces introduced the ability to manage memory exclusive to teams without the need for global synchronisation. However, one problem still remains that affects the usability of teams, and that is the need for the user to manage memory used internally by the implementation for synchronisation in collective operations. This paper explores the possibilities for moving this responsibility from the user to the implementation, as well as the consequences that may arise as a result. To this end, we describe three methods of implementation and discuss the implications of their use compared to traditional user management of synchronisation buffers.
- Research Organization:
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)
- Sponsoring Organization:
- USDOE Office of Science (SC)
- DOE Contract Number:
- AC05-00OR22725
- OSTI ID:
- 1567412
- Journal Information:
- OPENSHMEM AND RELATED TECHNOLOGIES: ENHANCING OPENSHMEM FOR HYBRID ENVIRONMENTS, Vol. 10007; Conference: Workshop on OpenSHMEM and Related Technologies, Baltimore, MD, August 2-4, 2016
- Country of Publication:
- United States
- Language:
- English
Extending the OpenSHMEM Memory Model to Support User-Defined Spaces
|
conference | January 2014 |
An evaluation of directory schemes for cache coherence
|
journal | May 1988 |
A low-overhead coherence solution for multiprocessors with private cache memories
|
journal | June 1984 |
Analysis of Programs for Parallel Processing
|
journal | October 1966 |
DeNovo: Rethinking the Memory Hierarchy for Disciplined Parallelism
|
conference | October 2011 |
A composite and scalable cache coherence protocol for large scale CMPs
|
conference | January 2011 |
The Scalable Heterogeneous Computing (SHOC) benchmark suite
|
conference | January 2010 |
Similar Records
Integrating Asynchronous Task Parallelism with OpenSHMEM, In: OpenSHMEM 2016: OpenSHMEM and Related Technologies. Enhancing OpenSHMEM for Hybrid Environments
Designing a High Performance OpenSHMEM Implementation using Universal Common Communication Substrate as a Communication Middleware