Distributed-optimization-based mix- flow scheduling mechanism for data center networks

被引：0
|
作者：
Zhang T. [1 ,2 ]
Ren F. [3 ]
Shu R. [4 ]
机构：
[1] College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing
[2] Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing
[3] Department of Computer Science and Technology, Tsinghua University, Beijing
[4] Microsoft Research, Beijing
来源：
Qinghua Daxue Xuebao/Journal of Tsinghua University | 2021年 / 61卷 / 06期
关键词：
Data center network (DCN); Deadline miss rate (DMR); Distributed optimization; Flow completion time (FCT); Mix-flow scheduling;
D O I：
10.16511/j.cnki.qhdxxb.2020.21.018
中图分类号：
学科分类号：
摘要：
Data center networks are key cloud computing infrastructure whose performance critically impacts the quality of service. Currently, data centers have multiple services with both deadline and non-deadline flows. This paper presents a distributed-optimization-based mix-flow scheduling (DOMS) mechanism to meet the transmission requirements of both types of flows. First, the optimization goals and transmission constraints are defined for both kinds of flows and the mixed-flow scheduling problem is formalized as a real-time rate allocation problem. Then, a coordinated scheduling structure is designed for the hosts and switches that leverages the dual decomposition characteristics of the problem. This method uses a distributed solution method to solve the problem with the flow rates evolving to a global optimal solution. Simulations show that this method effectively reduces deadline miss rates for deadline flows as well as flow completion times for non-deadline flows. © 2021, Tsinghua University Press. All right reserved.
引用
收藏
页码：618 / 625
页数：7
相关论文
共 16 条
  • [1] Cisco, Cisco global cloud index: Forecast and methodology, 2016-2021, (2018)
  • [2] NOORMOHAMMADPOUR M, RAGHAVENDRA C S., Datacenter traffic control: Understanding techniques and tradeoffs, IEEE Communications Surveys and Tutorials, 20, 2, pp. 1492-1525, (2018)
  • [3] CHEN L, CHEN K, BAI W, Et al., Scheduling mix-flows in commodity datacenters with Karuna, Proceedings of the ACM SIGCOMM 2016 Conference, pp. 174-187, (2016)
  • [4] WILSON C, BALLANI H, KARAGIANNIS T, Et al., Better never than late: Meeting deadlines in datacenter networks, Proceedings of the ACM SIGCOMM 2011 Conference on Applications, Technologies, Architectures, and Protocols for Computer Communications, pp. 50-61, (2011)
  • [5] CHOWDHURY M, ZHONG Y, STOICA I., Efficient coflow scheduling with Varys, ACM SIGCOMM 2014 Conference, pp. 443-454, (2014)
  • [6] HONG C Y, CAESAR M, GODFREY P B., Finishing flows quickly with preemptive scheduling, ACM SIGCOMM 2012 Conference, pp. 127-138, (2012)
  • [7] ALIZADEH M, YANG S, SHARIF M, Et al., pFabric: Minimal near-optimal datacenter transport, ACM SIGCOMM 2013 Conference, pp. 435-446, (2013)
  • [8] PERRY J, OUSTERHOUT A, BALAKRISHNAN H, Et al., Fastpass: A centralized 'zero-queue' datacenter network, ACM SIGCOMM 2014 Conference, pp. 307-318, (2014)
  • [9] MUNIR A, BAIG G, IRTEZA S M, Et al., Friends, not foes: Synthesizing existing transport strategies for data center networks, ACM SIGCOMM 2014 Conference, pp. 491-502, (2014)
  • [10] BAI W, CHEN L, CHEN K, Et al., PIAS: Practical information-agnostic flow scheduling for commodity data centers, IEEE/ACM Transactions on Networking, 25, 4, pp. 1954-1967, (2017)
12