TY - JOUR
T1 - Datacenter congestion control
T2 - Identifying what is essential and making it practical
AU - Mushtaq, Aisha
AU - Alizadeh, Mohammad
AU - Mittal, Radhika
AU - Ratnasamy, Sylvia
AU - McCauley, James
AU - Shenker, Scott
N1 - Publisher Copyright:
© 2019 Association for Computing Machinery. All rights reserved.
PY - 2019/7
Y1 - 2019/7
N2 - Recent years have seen a slew of papers on datacenter congestion control mechanisms. In this editorial, we ask whether the bulk of this research is needed for the common case where congestion control involves hosts responding to simple congestion signals from the network and the performance goal is reducing some average measure of flow completion time. We raise this question because we find that, out of all the possible variations one could make in congestion control algorithms, the most essential feature is the switch scheduling algorithm. More specifically, we find that congestion control mechanisms that use Shortest-Remaining-Processing-Time (SRPT) achieve superior performance as long as the rate-setting algorithm at the host is reasonable. We further find that while SRPT's performance is quite robust to host behaviors, the performance of schemes that use scheduling algorithms like FIFO or Fair Queuing depend far more crucially on the rate-setting algorithm, and their performance is typically worse than what can be achieved with SRPT. Given these findings, we then ask whether it is practical to realize SRPT in switches without requiring custom hardware. We observe that approximate and deployable SRPT (ADS) designs exist, which leverage the small number of priority queues supported in almost all commodity switches, and require only software changes in the host and the switches. Our evaluations with one very simple ADS design shows that it can achieve performance close to true SRPT and is significantly better than FIFO. Thus, the answer to our basic question - whether the bulk of recent research on datacenter congestion control algorithms is needed for the common case - is no.
AB - Recent years have seen a slew of papers on datacenter congestion control mechanisms. In this editorial, we ask whether the bulk of this research is needed for the common case where congestion control involves hosts responding to simple congestion signals from the network and the performance goal is reducing some average measure of flow completion time. We raise this question because we find that, out of all the possible variations one could make in congestion control algorithms, the most essential feature is the switch scheduling algorithm. More specifically, we find that congestion control mechanisms that use Shortest-Remaining-Processing-Time (SRPT) achieve superior performance as long as the rate-setting algorithm at the host is reasonable. We further find that while SRPT's performance is quite robust to host behaviors, the performance of schemes that use scheduling algorithms like FIFO or Fair Queuing depend far more crucially on the rate-setting algorithm, and their performance is typically worse than what can be achieved with SRPT. Given these findings, we then ask whether it is practical to realize SRPT in switches without requiring custom hardware. We observe that approximate and deployable SRPT (ADS) designs exist, which leverage the small number of priority queues supported in almost all commodity switches, and require only software changes in the host and the switches. Our evaluations with one very simple ADS design shows that it can achieve performance close to true SRPT and is significantly better than FIFO. Thus, the answer to our basic question - whether the bulk of recent research on datacenter congestion control algorithms is needed for the common case - is no.
KW - Congestion Control
KW - SRPT
KW - TCP
UR - http://www.scopus.com/inward/record.url?scp=85075079257&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075079257&partnerID=8YFLogxK
U2 - 10.1145/3371927.3371932
DO - 10.1145/3371927.3371932
M3 - Article
AN - SCOPUS:85075079257
SN - 0146-4833
VL - 49
SP - 32
EP - 38
JO - Computer Communication Review
JF - Computer Communication Review
IS - 3
ER -