Abstract / Description of output
In datacenters, workload throughput is often constrained by the attachment bandwidth of proxy servers, despite the much higher aggregate bandwidth of backend servers. We introduce a novel architecture that addresses this problem by combining programmable network switches with a controller that together act as a network "Prism" that can transparently redirect individual client transactions to diferent backend servers. Unlike traditional proxy approaches, with Prism, transaction payload data is exchanged directly between clients and backend servers, which eliminates the proxy bottleneck. Because the controller only handles transactional metadata, it should scale to much higher transaction rates than traditional proxies. An experimental evaluation with a prototype implementation demonstrates correctness of operation, improved bandwidth utilization and low packet transformation overheads even in software.
Original language | English |
---|---|
Title of host publication | Proceedings of the 2017 Symposium on Cloud Computing |
Place of Publication | New York, NY, USA |
Publisher | ACM Association for Computing Machinery |
Pages | 181–188 |
Number of pages | 8 |
ISBN (Print) | 9781450350280 |
DOIs | |
Publication status | Published - 24 Sept 2017 |
Event | ACM Symposium on Cloud Computing 2017 - Santa Clara, United States Duration: 25 Sept 2017 → 27 Sept 2017 https://acmsocc.github.io/2017/index.html |
Symposium
Symposium | ACM Symposium on Cloud Computing 2017 |
---|---|
Abbreviated title | SoCC '17 |
Country/Territory | United States |
City | Santa Clara |
Period | 25/09/17 → 27/09/17 |
Internet address |
Keywords / Materials (for Non-textual outputs)
- load-balancing
- proxying
- rewriting
- TCP
- datacenter