Abstract
The cache coherence protocol plays an important role in the performance of distributed and centralized shared-memory multiprocessors. A variety of bus-based cache coherence protocols exist and differ mainly in the way they respond to the transactions, and the bus transition state. These protocols can be complex and their impact on the performance of a multiprocessor system is often difficult to assess. In this paper, we present an Improved-MOESI cache coherence protocol. To measure the performance of the Improved-MOESI protocol, an existing simulator is modified and ported and a trace format converter program is written. The proposed Improved-MOESI, classic MOESI, MESI and MSI cache coherence protocols are implemented and simulated. The results show that the overall performance of the Improved-MOESI is better than the classic MOESI, MSI and MESI cache coherence protocols.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
