skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Simulation analysis of data sharing in shared-memory multiprocessors

Miscellaneous ·
OSTI ID:6155317

This dissertation examines shared memory reference patterns in parallel programs that run on bus-based, shared memory multiprocessors. The study reveals two distinct modes of sharing behavior. In sequential sharing, a processor makes multiple, sequential writes to the words within a block, uninterrupted by accesses from other processors. Under fine-grain sharing, processors contend for these words, and the number of per-processor sequential writes is low. Whether a program exhibits sequential of fine-grain sharing affects several factors relating to multiprocessor performance: the accuracy of sharing models that predict cache coherency overhead, the cache miss ratio and bus utilization of parallel programs, and the choice of coherency protocol. An architecture-independent model of write sharing was developed, based on the inter-processor activity to write-shared data. The model was used to predict the relative coherency overhead of write-invalidate and write-broadcast protocols. Architecturally detailed simulations validated the model for write-broadcast. Successive refinements, incorporating architecture-dependent parameters, most importantly cache block size, produced acceptable predictions for write-invalidate. Block size was crucial for modeling write-invalidate, because the pattern of memory references within a block determines protocol performance. The cache and bus behavior of parallel programs running under write-invalidate protocols was evaluated over various block and cache sizes. The analysis determined the effect of shared memory accesses on cache miss ratio and bus utilization by focusing on the sharing component of these metrics. The studies show that parallel programs incur substantially higher miss ratios and bus utilization than comparable uniprocessor programs. The sharing component of the metrics proportionally increases with cache and block size.

Research Organization:
California Univ., Berkeley, CA (USA)
OSTI ID:
6155317
Resource Relation:
Other Information: Thesis (Ph.D)
Country of Publication:
United States
Language:
English