The high performance Fortran handbook
The high performance Fortran handbook
Compiling high performance Fortran for distributed-memory systems
Digital Technical Journal
Design of Digital's parallel software environment
Digital Technical Journal
Building a high-performance message-passing system for MEMORY CHANNEL clusters
Digital Technical Journal
Experience with Parallel Computing on the AN2 Network
IPPS '96 Proceedings of the 10th International Parallel Processing Symposium
Digital's clusters and scientific parallel applications
COMPCON '96 Proceedings of the 41st IEEE International Computer Conference
Overview of memory channel network for PCI
COMPCON '96 Proceedings of the 41st IEEE International Computer Conference
The Performance of PVM on MPP Systems
The Performance of PVM on MPP Systems
The Virtual Interface Architecture
IEEE Micro
The Implementation of Low Latency Communication Primitives in the Snow Prototype
ICPP '97 Proceedings of the international Conference on Parallel Processing
Partitioning and scheduling loops on NOWs
Computer Communications
| Hi-index | 0.00 |
Historically, cluster systems have been connected with standard networking technology. In this presentation, we discuss the experience of using a highly optimized network, Memory ChannelTM for PCI, to enhance both the parallel performance and high-availability aspects of a cluster. Performance results for various approaches to fast messaging will be presented. These results will be compared to the same systems using conventional network technology. Finally, we briefly discuss our short-term futures for Memory Channel.