Sciweavers

5640 search results - page 982 / 1128
» The Primacy of Process Architecture
Sort
View
IPPS
2000
IEEE
14 years 2 months ago
Using Switch Directories to Speed Up Cache-to-Cache Transfers in CC-NUMA Multiprocessors
In this paper, we propose a novel hardware caching technique, called switch directory, to reduce the communication latency in CC-NUMA multiprocessors. The main idea is to implemen...
Ravi R. Iyer, Laxmi N. Bhuyan, Ashwini K. Nanda
IPPS
2000
IEEE
14 years 2 months ago
Controlling Distributed Shared Memory Consistency from High Level Programming Languages
One of the keys for the success of parallel processing is the availability of high-level programming languages for on-the-shelf parallel architectures. Using explicit message passi...
Yvon Jégou
IPPS
2000
IEEE
14 years 2 months ago
Reducing Ownership Overhead for Load-Store Sequences in Cache-Coherent Multiprocessors
Parallel programs that modify shared data in a cachecoherent multiprocessor with a write-invalidate coherence protocol create ownership overhead in the form of ownership acquisiti...
Jim Nilsson, Fredrik Dahlgren
IPPS
2000
IEEE
14 years 2 months ago
Ordering Unstructured Meshes for Sparse Matrix Computations on Leading Parallel Systems
Abstract. Computer simulations of realistic applications usually require solving a set of non-linear partial di erential equations PDEs over a nite region. The process of obtaini...
Leonid Oliker, Xiaoye S. Li, Gerd Heber, Rupak Bis...
ISSRE
2000
IEEE
14 years 2 months ago
Software Black Box: An Alternative Mechanism for Failure Analysis
Learning from software failures is an essential step towards the development of more reliable software systems and processes. However, as more intricate software systems are devel...
Sebastian G. Elbaum, John C. Munson