TY - GEN
T1 - TIMELY
T2 - ACM Conference on Special Interest Group on Data Communication, SIGCOMM 2015
AU - Mittal, Radhika
AU - Lam, Vinh The
AU - Dukkipati, Nandita
AU - Blem, Emily
AU - Wassel, Hassan
AU - Ghobadi, Monia
AU - Vahdat, Amin
AU - Wang, Yaogong
AU - Wetherall, David
AU - Zats, David
N1 - Publisher Copyright:
© 2015 ACM.
Copyright:
Copyright 2017 Elsevier B.V., All rights reserved.
PY - 2015/8/17
Y1 - 2015/8/17
N2 - Datacenter transports aim to deliver low latency messaging together with high throughput. We show that simple packet delay, measured as round-trip times at hosts, is an effective congestion signal without the need for switch feedback. First, we show that advances in NIC hardware have made RTT measurement possible with microsecond accuracy, and that these RTTs are sufficient to estimate switch queueing. Then we describe how TIMELY can adjust transmission rates using RTT gradients to keep packet latency low while delivering high bandwidth. We implement our design in host software running over NICs with OS-bypass capabilities. We show using experiments with up to hundreds of machines on a Clos network topology that it provides excellent performance: Turning on TIMELY for OS-bypass messaging over a fabric with PFC lowers 99 percentile tail latency by 9X while maintaining near line-rate throughput. Our system also outperforms DCTCP running in an optimized kernel, reducing tail latency by 13X. To the best of our knowledge, TIMELY is the first delay-based congestion control protocol for use in the datacenter, and it achieves its results despite having an order of magnitude fewer RTT signals (due to NIC offload) than earlier delay-based schemes such as Vegas.
AB - Datacenter transports aim to deliver low latency messaging together with high throughput. We show that simple packet delay, measured as round-trip times at hosts, is an effective congestion signal without the need for switch feedback. First, we show that advances in NIC hardware have made RTT measurement possible with microsecond accuracy, and that these RTTs are sufficient to estimate switch queueing. Then we describe how TIMELY can adjust transmission rates using RTT gradients to keep packet latency low while delivering high bandwidth. We implement our design in host software running over NICs with OS-bypass capabilities. We show using experiments with up to hundreds of machines on a Clos network topology that it provides excellent performance: Turning on TIMELY for OS-bypass messaging over a fabric with PFC lowers 99 percentile tail latency by 9X while maintaining near line-rate throughput. Our system also outperforms DCTCP running in an optimized kernel, reducing tail latency by 13X. To the best of our knowledge, TIMELY is the first delay-based congestion control protocol for use in the datacenter, and it achieves its results despite having an order of magnitude fewer RTT signals (due to NIC offload) than earlier delay-based schemes such as Vegas.
KW - Datacenter transport
KW - Delay-based congestion control
KW - Osbypass
KW - Rdma
UR - http://www.scopus.com/inward/record.url?scp=84962215398&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84962215398&partnerID=8YFLogxK
U2 - 10.1145/2785956.2787510
DO - 10.1145/2785956.2787510
M3 - Conference contribution
AN - SCOPUS:84962215398
T3 - SIGCOMM 2015 - Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication
SP - 537
EP - 550
BT - SIGCOMM 2015 - Proceedings of the 2015 ACM Conference on Special Interest Group on Data Communication
PB - Association for Computing Machinery, Inc
Y2 - 17 August 2015 through 21 August 2015
ER -