On Application-level Approaches to Avoiding TCP Throughput Collapse in Cluster-based Storage Systems
- Elie Krevat ,
- Vijay Vasudevan ,
- Amar Phanishayee ,
- David Andersen ,
- Gregory Ganger ,
- Garth Gibson ,
- Srinivasan Seshan
ACM Petascale Data Storage Workshop |
TCP Incast plagues scalable cluster-based storage built atop standard TCP/IP-over-Ethernet, often resulting in much lower client read bandwidth than can be provided by the available network links. This paper reviews the Incast problem and discusses potential application-level approaches to avoiding it.