2016
DOI: 10.7717/peerj-cs.98
|View full text |Cite
|
Sign up to set email alerts
|

SNF: synthesizing high performance NFV service chains

Abstract: In this paper we introduce SNF, a framework that synthesizes (S) network function (NF) service chains by eliminating redundant I/O and repeated elements, while consolidating stateful cross layer packet operations across the chain. SNF uses graph composition and set theory to determine traffic classes handled by a service chain composed of multiple elements. It then synthesizes each traffic class using a minimal set of new elements that apply single-read-single-write and early-discard operations. Our SNF protot… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

0
20
0

Year Published

2017
2017
2022
2022

Publication Types

Select...
5
4

Relationship

2
7

Authors

Journals

citations
Cited by 35 publications
(20 citation statements)
references
References 13 publications
0
20
0
Order By: Relevance
“…(3) Scheduling decisions are made at packet-level vs flow-level, making our solution more adaptive to traffic dynamics. Some recent works [31], [49]- [51] consider optimizations and scheduling at the level of a single server or CPU core. In particular, NFVnice is a VNF framework for CPUs that aims for fair and efficient resource allocation of chains, considering the impact of different VNFs on resource usage.…”
Section: Related Workmentioning
confidence: 99%
“…(3) Scheduling decisions are made at packet-level vs flow-level, making our solution more adaptive to traffic dynamics. Some recent works [31], [49]- [51] consider optimizations and scheduling at the level of a single server or CPU core. In particular, NFVnice is a VNF framework for CPUs that aims for fair and efficient resource allocation of chains, considering the impact of different VNFs on resource usage.…”
Section: Related Workmentioning
confidence: 99%
“…In addition to user-space I/O frameworks, there have also been efforts to optimize the kernel network stack [18,22,59]. Additionally, several efforts have been made to improve NFV application performance when running over user-space I/O frameworks [8,16,33,34,62,68,86]. Furthermore, there are a limited number of works that employed CAT to mitigate noisy neighbor effect and improve NFV performance [70,72,73].…”
Section: Related Workmentioning
confidence: 99%
“…Metron [25] translates service chains' traffic classes (synthesized by SNF [26]) to NIC or OpenFlow classification rules (see Figure 4c). We refer to this method as traffic-class based in Table 1 and Figure 4b.…”
Section: Intra-server Load-balancing Techniquesmentioning
confidence: 99%
“…low latency. Recent developments in key-value stores [9,21,22,33,39], Network Functions Virtualization (NFV) [25,26], network stacks [12,19,49,60], and high-speed networking platforms [4,15,48] all advocate the use of sharding. Sharding divides resources (e.g., CPU cores and memory) into multiple shards that process requests in parallel, in a totally independent way.…”
mentioning
confidence: 99%