Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Traffic Characteristics on 1Gbit/s Access Aggregation Links
RISE - Research Institutes of Sweden, ICT, SICS. (Decisions, Networks and Analytics lab)ORCID-id: 0000-0002-8102-5773
RISE - Research Institutes of Sweden, ICT, SICS. (Decisions, Networks and Analytics lab)ORCID-id: 0000-0002-1445-2169
TeliaSonera, Sweden.
TeliaSonera, Sweden.
Visa övriga samt affilieringar
2017 (Engelska)Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

Large network operators have thousands or tens of thousands of access aggregation links that they need to manage and dimension properly. Measuring and understanding the traffic characteristics on these type of links are therefore essential. What do the traffic intensity characteristics look like on different timescales from days down to milliseconds? How do the characteristics differ if we compare links with the same capacity but with different type of clients and access technologies? How do the traffic characteristics differ from that on core network links? These are the type of questions we set out to investigate in this paper. We present the results of packet level measurements on three different 1Gbit/s aggregation links in an operational IP network. We see large differences in traffic characteristics between the three links. We observe highly skewed link load probability densities on timescales relevant for buffering (i.e. 10-milliseconds). We demonstrate the existence of large traffic spikes on short timescales (10-100ms) and show their impact on link delay. We also found that these traffic bursts often are caused by only one or a few IP flows.

Ort, förlag, år, upplaga, sidor
2017.
Nationell ämneskategori
Data- och informationsvetenskap
Identifikatorer
URN: urn:nbn:se:ri:diva-32376OAI: oai:DiVA.org:ri-32376DiVA: diva2:1152734
Konferens
IEEE International Conference on Communications (ICC), 21-25 May 2017, Paris, France
Anmärkning

Funder: The Knowledge Foundation (KKS)

Tillgänglig från: 2017-10-26 Skapad: 2017-10-26 Senast uppdaterad: 2017-11-20Bibliografiskt granskad

Open Access i DiVA

Fulltext saknas

Sök vidare i DiVA

Av författaren/redaktören
Abrahamsson, HenrikAhlgren, Bengt
Av organisationen
SICS
Data- och informationsvetenskap

Sök vidare utanför DiVA

GoogleGoogle Scholar

Totalt: 4 träffar
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf