Demonstration venue · read-only. Every page can be browsed; the buttons that would change it are switched off. Create an account to run TaxoReview on your own data.

Sudhur Ahuja

dblp:167/7200 · DBLP profile ↗
← Back
1ranked-venue papers
1as first author
0since 2021 · last 1983
—ORCID · none

Domains — the database's venue-derived domains; a paper can count in several

Computer networks · 1 · 1 first-author

Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.

Computer architecture, parallel and distributed computing, and storage systems
1 paper
Interconnection networks and networks-on-chip · 92% Distributed systems · 8%

Topics — the 5 heaviest of 5, each with the papers that count most for it

TopicWeightPapersLastEvidence papers
Interconnection networks and networks-on-chip
cluster interconnect
0.011983
S/NET: A High-Speed Interconnect for Multiple Computers · IEEE J. Sel. Areas Commun. 1983
Interconnection networks and networks-on-chip
high-speed interconnect
0.011983
S/NET: A High-Speed Interconnect for Multiple Computers · IEEE J. Sel. Areas Commun. 1983
Interconnection networks and networks-on-chip
switch architecture
0.011983
S/NET: A High-Speed Interconnect for Multiple Computers · IEEE J. Sel. Areas Commun. 1983
Distributed systems › operating system support › interprocess communication
message-passing communication
0.011983
S/NET: A High-Speed Interconnect for Multiple Computers · IEEE J. Sel. Areas Commun. 1983
Interconnection networks and networks-on-chip
multiprocessor interconnection
0.011983
S/NET: A High-Speed Interconnect for Multiple Computers · IEEE J. Sel. Areas Commun. 1983

Methods — techniques the papers use, named apart from their topics

performance evaluation · 0.0
YearPublicationVenuePosition
1983 S/NET: A High-Speed Interconnect for Multiple Computers
abstract
This paper describes S/NET (symmetric network), a high-speed small area interconnect that supports effective multiprocessing using message-based communication. This interconnect provides low latency, bounded contention time, and high throughput. It further provides hardware support for low level flow control and signaling. The interconnect is a star network with an active switch. The computers connect to the switch through full duplex fiber links. The S/NET provides a simple memory addressable interface to the processors and appears as a logical bus interconnect. The switch provides fast, fair, and deterministic contention resolution. It further supports high priority signals to be sent unimpeded in presence of data traffic (this can viewed as equivalent to interrupts on a conventional memory bus). The initial implementation supports a mix of VAX computers and Motorola 68000 based single board computers up to a maximum of 12. The switch throughput is 80 Mbits/s and the fiber links operate at a data rate of 10 Mbits/s. The kernel-to-kernel latency is only100 \mus. We present a description of the architecture and discuss the performance of current systems.
Sudhur Ahuja
IEEE J. Sel. Areas Commun.1