VM-based shared memory on low-latency, remote-memory-access networks

• Published in: Conference Proceedings. The 24th Annual International Symposium on Computer Architecture pp. 157 - 169
• Main Authors: Kontothanassis, Leonidas, Hunt, Galen, Stets, Robert, Hardavellas, Nikolaos, Cierniak, Michał, Parthasarathy, Srinivasan, Meira, Wagner, Dwarkadas, Sandhya, Scott, Michael
• Format: Conference Proceeding
• Language: English
• Published: New York, NY, USA ACM 01.01.1997; IEEE
• Series: ACM Conferences
• Subjects: Access protocols; Broadcasting; Computer science; Computer systems organization > Dependable and fault-tolerant systems and networks; Contracts; Delay; General and reference > Cross-computing tools and techniques > Performance; Hardware; Hardware > Communication hardware, interfaces and storage; Networks; Networks > Network performance evaluation; Permission; Software performance; Space technology; Workstations
• ISBN: 9780897919012; 0897919017
• ISSN: 1063-6897; 2575-713X
• DOI: 10.1145/264107.264163

Recent technological advances have produced network interfaces that provide users with very low-latency access to the memory of remote machines. We examine the impact of such networks on the implementation and performance of software DSM. Specifically, we compare two DSM systems---Cashmere and TreadMarks---on a 32-processor DEC Alpha cluster connected by a Memory Channel network.Both Cashmere and TreadMarks use virtual memory to maintain coherence on pages, and both use lazy, multi-writer release consistency. The systems differ dramatically, however, in the mechanisms used to track sharing information and to collect and merge concurrent updates to a page, with the result that Cashmere communicates much more frequently, and at a much finer grain.Our principal conclusion is that low-latency networks make DSM based on fine-grain communication competitive with more coarse-grain approaches, but that further hardware improvements will be needed before such systems can provide consistently superior performance. In our experiments, Cashmere scales slightly better than TreadMarks for applications with false sharing. At the same time, it is severely constrained by limitations of the current Memory Channel hardware. In general, performance is better for TreadMarks.