Network Function Offloading on Programmable Switch with Sequential Pipelines

Network Function Virtualization (NFV) allows network functions (NFs) to be chained as a service function chain (SFC) and deployed on a number of servers through virtualization technology, enabling flexible processing of network functions. With the development of programmable switches in recent years, some switches, such as P4 switches, are also capable of processing NFs, which can effectively improve the throughput of the system and reduce the processing latency of SFCs. But current programmable switches can only allow packets to be processed in a fixed NF order. Such a feature creates new problems in situations where network SFC requests are in a dynamic randomized manner. Based on the above background, this paper proposes a policy algorithm for accelerating SFC processing based on P4 switches. Our scheme take into account the sequential requirements of SFCs and the sequential structure of the P4 switch pipeline, and uses multiple switches to jointly process SFCs, which is able to handle dynamically generated SFCs with superior performance. Compared with other algorithms, our algorithm is able to achieve lower latency.

Authors

Institutions

Publication Details

Published
2025-03-14
DOI
https://doi.org/10.1109/itoec63606.2025.10967581
Primary Topic
Software-Defined Networks and 5G
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Network Function Offloading on Programmable Switch with Sequential Pipelines

Deyun Gao, Dong Yang, Yan Chen, Bo Pang
Software-Defined Networks and 5G
article

Network Function Offloading on Programmable Switch with Sequential Pipelines

Deyun Gao, Dong Yang, Yan Chen, Bo Pang
article en

Abstract

Network Function Virtualization (NFV) allows network functions (NFs) to be chained as a service function chain (SFC) and deployed on a number of servers through virtualization technology, enabling flexible processing of network functions. With the development of programmable switches in recent years, some switches, such as P4 switches, are also capable of processing NFs, which can effectively improve the throughput of the system and reduce the processing latency of SFCs. But current programmable switches can only allow packets to be processed in a fixed NF order. Such a feature creates new problems in situations where network SFC requests are in a dynamic randomized manner. Based on the above background, this paper proposes a policy algorithm for accelerating SFC processing based on P4 switches. Our scheme take into account the sequential requirements of SFCs and the sequential structure of the P4 switch pipeline, and uses multiple switches to jointly process SFCs, which is able to handle dynamically generated SFCs with superior performance. Compared with other algorithms, our algorithm is able to achieve lower latency.

Beijing Jiaotong University (CN)
National Natural Science Foundation of China
Affordable and clean energy
Openalex Percentile: Top 92%
Software-Defined Networks and 5G
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.