Message passing vs. shared address space on a cluster of SMPs
- 13 November 2002
- proceedings article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
The emergence of scalable computer architectures using clusters of PCs (or PC-SMPs) with commodity networking has made them attractive platforms for high-end scientific computing. Currently, message passing (MP) and shared address space (SAS) are the two leading programming paradigms for these systems. MP has been standardized with MPI, and is the most common and mature parallel programming approach. However, MP code development can be extremely difficult, especially for irregularly structured computations. SAS offers substantial ease of programming, but may suffer from performance limitations due to poor spatial locality and high protocol overhead. In this paper, we compare the performance of and programming effort required for six applications under both programming models on a 32-CPU PC-SMP cluster. Our application suite consists of codes that typically do not exhibit scalable performance under shared-memory programming due to their high communication-to-computation ratios and complex communication patterns. Results indicate that SAS can achieve about half the parallel efficiency of MPI for most of our applications; however on certain classes of problems, SAS performance is competitive with MPIKeywords
This publication has 11 references indexed in Scilit:
- Using network interface support to avoid asynchronous protocol processing in shared virtual memory systemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Tempest and Typhoon: user-level shared memoryPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Parallel tree building on a range of shared address space multiprocessors: algorithms and application performancePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Application scaling under shared virtual memory on a cluster of SMPsPublished by Association for Computing Machinery (ACM) ,1999
- Scaling application performance on a cache-coherent multiprocessorACM SIGARCH Computer Architecture News, 1999
- Quantifying the Performance Differences between PVM and TreadMarksJournal of Parallel and Distributed Computing, 1997
- ShastaPublished by Association for Computing Machinery (ACM) ,1996
- Myrinet: a gigabit-per-second local area networkIEEE Micro, 1995
- Fine-grain access control for distributed shared memoryPublished by Association for Computing Machinery (ACM) ,1994
- Memory coherence in shared virtual memory systemsACM Transactions on Computer Systems, 1989