Flow-Aware Service Function Embedding Algorithm in Programmable Data Plane

Jaewook Lee, Haneul Ko, Hochan Lee, Sangheon Pack

Research output: Contribution to journalArticlepeer-review

Abstract

Service function chaining (SFC) is an indispensable technique for Internet service providers to efficiently manage their networks. However, SFC poses requirements of additional processing time of service functions (SFs) and increased routing time owing to detoured paths. In this paper, we introduce the use of a programmable data plane (PDP) to reduce the additional processing and routing times in SFC. We first classify the existing PDP-empowered SFC schemes and analyze their pros and cons. An optimization problem, to find the optimal SF embedding strategy minimizing the SFC completion time while efficiently utilizing the PDP switch resources, is formulated and a flow-aware SF embedding (FASE) algorithm that complementarily combines the redundant SF and re-circulation approaches is devised. FASE is implemented over a commercial PDP switch and experimental results demonstrate that FASE can reduce the SFC completion time by up to 33% compared with conventional approaches while utilizing the switch resources efficiently.

Original languageEnglish
Article number9311727
Pages (from-to)6113-6121
Number of pages9
JournalIEEE Access
Volume9
DOIs
Publication statusPublished - 2021

Keywords

  • Service function chaining
  • in-network computing
  • programmable data plane

ASJC Scopus subject areas

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)

Fingerprint Dive into the research topics of 'Flow-Aware Service Function Embedding Algorithm in Programmable Data Plane'. Together they form a unique fingerprint.

Cite this