Name
Abstract | ||
|---|---|---|
This paper presents a practical solution to a problem facing high-fan-in, high-bandwidth synchronized TCP workloads in datacenter Ethernets---the TCP incast problem. In these networks, receivers can experience a drastic reduction in application throughput when simultaneously requesting data from many servers using TCP. Inbound data overfills small switch buffers, leading to TCP timeouts lasting hundreds of milliseconds. For many datacenter workloads that have a barrier synchronization requirement (e.g., filesystem reads and parallel data-intensive queries), throughput is reduced by up to 90%. For latency-sensitive applications, TCP timeouts in the datacenter impose delays of hundreds of milliseconds in networks with round-trip-times in microseconds. Our practical solution uses high-resolution timers to enable microsecond-granularity TCP timeouts. We demonstrate that this technique is effective in avoiding TCP incast collapse in simulation and in real-world experiments. We show that eliminating the minimum retransmission timeout bound is safe for all environments, including the wide-area. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1145/1592568.1592604 | SIGCOMM |
Keywords | Field | DocType |
high resolution,throughput,performance,round trip time | H-TCP,TCP Westwood plus,Compound TCP,Computer science,Computer network,TCP acceleration,Zeta-TCP,TCP tuning,TCP global synchronization,TCP Friendly Rate Control,Distributed computing | Conference |
Volume | Issue | ISSN |
39 | 4 | 0146-4833 |
Citations | PageRank | References |
215 | 14.46 | 26 |
Authors | ||
8 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Vijay Vasudevan | 1 | 449 | 28.44 |
| Amar Phanishayee | 2 | 804 | 56.59 |
| Hiral Shah | 3 | 215 | 14.46 |
| Elie Krevat | 4 | 667 | 53.83 |
| andersen david g | 5 | 4823 | 345.31 |
| Gregory R. Ganger | 6 | 4560 | 383.16 |
| Garth A. Gibson | 7 | 2517 | 250.27 |
| Brian Mueller | 8 | 418 | 29.04 |