hh.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Fiber-ribbon ring network with services for parallel processing and distributed real-time systems
Halmstad University, School of Information Science, Computer and Electrical Engineering (IDE), Halmstad Embedded and Intelligent Systems Research (EIS), Embedded Systems (CERES).ORCID iD: 0000-0002-6526-3931
Halmstad University, School of Information Science, Computer and Electrical Engineering (IDE), Halmstad Embedded and Intelligent Systems Research (EIS), Embedded Systems (CERES).
Halmstad University, School of Information Science, Computer and Electrical Engineering (IDE), Halmstad Embedded and Intelligent Systems Research (EIS), Embedded Systems (CERES).
1999 (English)In: Proc. ISCA 12th International Conference on Parallel and Distributed Computing Systems (PDCS-99), ISCA , 1999, 94-101 p.Conference paper, (Refereed)
Abstract [en]

In this paper, we present how real-time services are implemented in a control-channel based ring network built up of fiber-ribbon point-to-point links. Services for best effort messages, guarantee seeking messages and real-time virtual channels are supported for single destination, multicast and broadcast transmission by the network. Slot-reservation is used for the implementation of real-time virtual channels. High aggregated throughput can be achieved due to pipelining, i.e., data can be transmitted simultaneously in different segments of the ring. An analysis of worst-case latency and deterministic throughput, which are important measures for real-time service implementation, is provided. The network is analyzed for two variants of time-slot organization, one that offers higher throughput and one that offers lower latency in some situations. We also show how the network offers low-level support for parallel computing, i.e.,  barrier-synchronization and global reduction. The control channel is used when realizing these functions, which implies no modification of the original network architecture. Low-level support for reliable transmission is also offered in a similar manner. This includes acknowledge / negative acknowledge and flow-control.

Place, publisher, year, edition, pages
ISCA , 1999. 94-101 p.
Keyword [en]
Real-time communication, fiber-optic communication
National Category
Computer Engineering
Identifiers
URN: urn:nbn:se:hh:diva-39OAI: oai:DiVA.org:hh-39DiVA: diva2:232224
Available from: 2009-08-20 Created: 2009-08-20 Last updated: 2014-01-23Bibliographically approved

Open Access in DiVA

fulltext(98 kB)314 downloads
File information
File name FULLTEXT01.pdfFile size 98 kBChecksum SHA-512
7023cdda1785df66a0cfee4b692bc69e1563d1932dbdfec84b8b5e102033176ff35e35cf63ab33d65cea8c8ef373d1d88609484071229c19bc4e18082beb9876
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Jonsson, Magnus
By organisation
Embedded Systems (CERES)
Computer Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 314 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 152 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf