2017 · 252 citations · 40 references
ComputingComputer ScienceComputer SystemsComputer NetworksInterconnection NetworksQueuing TheoryCongestion ControlCongestion ManagementFlow ControlQueueing NetworksQueueing SystemsQueueing TheoryCongestion AvoidanceCongestion Control In SwitchesCongestion Control ProtocolsConvergenceData Loss ToleranceDatacenter Congestion ControlDelay-based Congestion ControlInternet Congestion Control
Datacenter networks face congestion control challenges due to microsecond RTTs, bursty multi‑thousand concurrent flows, shallow buffers, and high link speeds that slow bandwidth probing. The goal is to design a congestion control algorithm that achieves zero data loss, fast convergence, low buffer occupancy, and high utilization.
Small RTTs (~tens of microseconds), bursty flow arrivals, and a large number of concurrent flows (thousands) in datacenters bring fundamental challenges to congestion control as they either force a flow to send at most one packet per RTT or induce a large queue build-up. The widespread use of shallow buffered switches also makes the problem more challenging with hosts generating many flows in bursts. In addition, as link speeds increase, algorithms that gradually probe for bandwidth take a long time to reach the fair-share. An ideal datacenter congestion control must provide 1) zero data loss, 2) fast convergence, 3) low buffer occupancy, and 4) high utilization. However, these requirements present conflicting goals.
40
Sanjay Ghemawat · Communications of the ACM · 2008 · 18.5K citations · Full text
Sangtae Ha, Injong Rhee, Lisong Xu · ACM SIGOPS Operating Systems Review · 2008 · 2.2K citations
Network traffic characteristics of data centers in the wild
Theophilus Benson, Aditya Akella, David A. Maltz · 2010 · 2.2K citations