A cost-comparison approach for adaptive distributed shared memory
ICS '96 Proceedings of the 10th international conference on Supercomputing
| Hi-index | 0.00 |
The focus of this report is on software implementations of Distributed Shared Memory (DSM). In recent years, many protocols for implementing DSM have been proposed. The protocols can be broadly divided into two classes: invalidation-based schemes and update-based schemes. Performance of the protocols depends on the memory access behavior of the applications. Some researchers have proposed DSMs that provide a family of consistency protocols or application specific protocols, and the programmer is allowed to choose any one of them for each shared memory object (or page) or each stage of an application. While such implementations have a potential for achieving optimal performance, they impose undue burden on the programmer. An adaptive implementation that automatically chooses the appropriate protocol for each shared memory page (at run-time) will ease the task of programming for DSM. This report presents a simple approach for implementing adaptive DSMs. The approach is illustrated with the example of an adaptive DSM based on the competitive update protocol. The objective of the adaptive scheme is to minimize a pre-defined "cost" function. The cost functions considered here are number of messages and size of messages. (Other cost functions can also be used similarly). The proposed scheme allows each node to independently choose (at run-time) a different protocol for each page. The report presents preliminary evaluation of the adaptive DSM. Preliminary results show that the performance is improved by dynamically selecting the appropriate protocol.