Active messages: a mechanism for integrated communication and computation
ISCA '92 Proceedings of the 19th annual international symposium on Computer architecture
Distributed network computing over local ATM networks
IEEE Journal on Selected Areas in Communications
U-Net: a user-level network interface for parallel and distributed computing
SOSP '95 Proceedings of the fifteenth ACM symposium on Operating systems principles
CRL: high-performance all-software distributed shared memory
SOSP '95 Proceedings of the fifteenth ACM symposium on Operating systems principles
Multicast virtual topologies for collective communication in MPCs and ATM clusters
Supercomputing '95 Proceedings of the 1995 ACM/IEEE conference on Supercomputing
Efficient support of location transparency in concurrent object-oriented programming languages
Supercomputing '95 Proceedings of the 1995 ACM/IEEE conference on Supercomputing
High performance messaging on workstations: Illinois fast messages (FM) for Myrinet
Supercomputing '95 Proceedings of the 1995 ACM/IEEE conference on Supercomputing
DPF: fast, flexible message demultiplexing using dynamic code generation
Conference proceedings on Applications, technologies, architectures, and protocols for computer communications
MBCF: a protected and virtualized high-speed user-level memory-based communication facility
ICS '98 Proceedings of the 12th international conference on Supercomputing
Shared Memory Programming in Metacomputing Environments: The Global Array Approach
The Journal of Supercomputing - Special issue: high performance distributed computing
A Performance Evaluation of the Convex SPP-1000 Scalable Shared Memory Parallel Computer
Supercomputing '95 Proceedings of the 1995 ACM/IEEE conference on Supercomputing
A Case for NOW (Networks of Workstations)
IEEE Micro
Assessing Fast Network Interfaces
IEEE Micro
Distributed computing using Java: a comparison of two server designs
Journal of Systems Architecture: the EUROMICRO Journal
The ParaStation project: Using workstations as building blocks for parallel computing
Information Sciences: an International Journal
| Hi-index | 0.00 |
Recent developments in communication architectures for parallel machines have reduced the communication overheads and latencies by over an order of magnitude. This paper examines whether these techniques can carry over to clusters of workstations connected by an ATM network even though clusters use standard operating system software, are equipped with network interfaces optimized for stream communication, do not allow direct protected user-level access to the network, and use networks without reliable transmission or flow control. In a first part, this paper describes the differences in communication characteristics between clusters of workstations built from standard hardware and software components and state-of-the-art multiprocessors. A second part evaluates a prototype implementation of the low-latency Active Messages communication model on a Sun workstation cluster interconnected by an ATM network. Measurements show application-level round-trip latencies of about 50 microseconds for small messages which is roughly comparable to the Active Messages implementation on the Thinking Machines CM-5 multiprocessor.