Distributed Shared Memory: A Survey of Issues and Algorithms
Computer - Distributed computing systems: separate resources acting as one
The SHRIMP performance monitor: design and applications
SPDT '96 Proceedings of the SIGMETRICS symposium on Parallel and distributed tools
Performance monitoring in a Myrinet-connected SHRIMP cluster
SPDT '98 Proceedings of the SIGMETRICS symposium on Parallel and distributed tools
SCI: Scalable Coherent Interface, Architecture and Software for High-Performance Compute Clusters
SCI: Scalable Coherent Interface, Architecture and Software for High-Performance Compute Clusters
SCI-VM: A Flexible Base for Transparent Shared Memory Programming Models on Clusters of PCs
Proceedings of the 11 IPPS/SPDP'99 Workshops Held in Conjunction with the 13th International Parallel Processing Symposium and 10th Symposium on Parallel and Distributed Processing
Supporting Shared Memory and Message Passing on Clusters of PCs with a SMiLE
CANPC '99 Proceedings of the Third International Workshop on Network-Based Parallel Computing: Communication, Architecture, and Applications
Optimizing Data Locality for SCI-Based PC-Clusters with the SMiLE Monitoring Approach
PACT '99 Proceedings of the 1999 International Conference on Parallel Architectures and Compilation Techniques
Brazos: a third generation DSM system
NT'97 Proceedings of the USENIX Windows NT Workshop on The USENIX Windows NT Workshop 1997
Hi-index | 0.00 |
Cost-effective clusters built from commodity-off-the-shelf components and connected with high-speed interconnection fabrics, together with easy-to-use shared memory programming models, are creating an attractive platform for parallel programming. However, these kinds of architectures currently lack monitoring environments that allow the observation of performance data at various levels, detection of bottlenecks, and overall optimization of applications. This work presents a comprehensive approach attacking the problem by combining three basic building blocks: monitoring hardware for a stateof-the-art system area network (SCI), an innovative hybrid distributed shared memory system providing the base for any kind of shared memory programming models (SCI Virtual Memory or SCI-VM), and an extensible online monitoring system (OMIS/OCM). This forms the basis for an extensive tool environment on top of this emerging platform, which allows easy application porting, debugging, and performance tuning.