Chaos Monkey: Increasing SDN Reliability through Systematic Network Destruction (Demo)
Conclusion
In this paper, we have lay the foundation for post-deployment failure injector for SDN Controllers. While this testing paradigm has been used in various cloud computing applications, this is the first step towards developing a solution for realtime SDN Controller verification. With many classes of network failure, it will be a continued effort to extend the SDN Chaos Monkey such that we can demonstrate coverage across possible failure classes. Developers of the controller platform will be able to use the SDN Chaos Monkey tool to test how the controller responds to any sequence of failures that could reasonably occur in the network. There are truly an endless amount of failures that could appear in the network, and it would be unreasonable for anyone – even Chaos Monkey – to cover them all. However, the Chaos Monkey utility encourages a testing paradigm that leads to continual and constant improvement on a complex piece of software upon which the entire network depends on. As the SDN controller becomes a more reliable and stable software agent, the SDN movement will gain further credibility and continue to disrupt the landscape of networking.
People
BibTex
@INPROCEEDINGS{chang2015chaos,
isbn = {978-1-4503-3542-3},
doi = {10.1145/2785956.2790038},
year = {2015},
booktitle = {Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication},
type = {Conference Paper},
author = {Chang, Michael A. and Tschaen, Brendan and Benson, Theophilus and Vanbever, Laurent},
language = {en},
address = {New York, NY},
publisher = {Association for Computing Machinery},
title = {Chaos Monkey: Increasing SDN Reliability through Systematic Network Destruction (Demo)},
PAGES = {371 - 372},
Note = {2015 ACM Conference on Special Interest Group on Data Communication (SIGCOMM 2015); Conference Location: London, United Kingdom; Conference Date: August 17-21, 2015}
}Research Collection: 20.500.11850/103896