Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
skip to main content
research-article

Large-scale parallel server system with multi-component jobs

Published: 01 June 2021 Publication History

Abstract

A broad class of parallel server systems is considered, for which we prove the steady-state asymptotic independence of server workloads, as the number of servers goes to infinity, while the system load remains sub-critical. Arriving jobs consist of multiple components. There are multiple job classes, and each class may be of one of two types, which determines the rule according to which the job components add workloads to the servers. The model is broad enough to include as special cases some popular queueing models with redundancy, such as cancel-on-start and cancel-on-completion redundancy. Our analysis uses mean-field process representation and the corresponding mean-field limits. In essence, our approach relies almost exclusively on three fundamental properties of the model: (a) monotonicity, (b) work conservation and (c) the property that, on average, “new arriving workload prefers to go to servers with lower workloads.”

References

[1]
Adan I, Kleiner I, Righter R, and Weiss G FCFS parallel service systems and matching models Perform. Eval. 2018 127 253-272
[2]
Ayesta U, Bodas T, and Verloop IM On a unifying product form framework for redundancy models Perform. Eval. 2018 127 93-119
[3]
Ayesta U, Bodas T, and Verloop IM On redundancy-d with cancel-on-start aka join-shortest-work (d) ACM SIGMETRICS Perform. Eval. Rev. 2018 46 2 24-26
[4]
Bramson M Stability of join the shortest queue networks Ann. Appl. Probab. 2011 21 1568-1625
[5]
Bramson M, Lu Y, and Prabhakar B Asymptotic independence of queues under randomized load balancing Queueing Syst. 2012 71 247-292
[6]
Foss S and Chernova N On the stability of a partially accessible multi-station queue with state-dependent routing Queueing Syst. 1998 29 55-73
[7]
Gardner K, Harchol-Balter M, Scheller-Wolf A, Velednitsky M, and Zbarsky S Redundancy-d: the power of d choices for redundancy Oper. Res. 2017 65 4 1078-1094
[8]
Gardner K, Zbarsky S, Doroudi S, Harchol-Balter M, and Hyytia E Reducing latency via redundant requests: exact analysis ACM SIGMETRICS Perform. Eval. Rev. 2015 43 1 347-360
[9]
Greenberg A, Malyshev V, and Popov S Stochastic model of massively parallel computation Markov Process. Relat. Fields 1997 2 473-490
[10]
Hellemans T, Bodas T, and Van Houdt B Performance analysis of workload dependent load balancing policies Proc. ACM Meas. Anal. Comput. Syst. 2019 3 2 35
[11]
Pang G, Talreja R, and Whitt W Martingale proofs of many-server heavy-traffic limits for Markovian queues Probab. Surv. 2007 4 193-267
[12]
Shah NB, Lee K, and Ramchandran K When do redundant requests reduce latency? IEEE Trans. Commun. 2015 64 2 715-722
[13]
Stolyar AL Pull-based load distribution in large-scale heterogeneous service systems Queueing Syst. 2015 80 4 341-361
[14]
Stolyar AL Pull-based load distribution among heterogeneous parallel servers: the case of multiple routers Queueing Syst. 2017 85 1–2 31-65
[15]
Vulimiri, A., Godfrey, P., Mittal, R., Sherry, J., Ratnasamy, S., Shenker, S.: Low latency via redundancy. CoNEXT 2013—Proceedings of the 2013 ACM International Conference on Emerging Networking Experiments and Technologies. Association for Computing Machinery, pp. 283–294 (2013)
[16]
Vvedenskaya N, Dobrushin R, and Karpelevich F Queueing system with selection of the shortest of two queues: an asymptotic approach Probl. Inf. Transm. 1996 32 1 20-34

Cited By

View all
  • (2024)Approximations to Study the Impact of the Service Discipline in Systems with RedundancyACM SIGMETRICS Performance Evaluation Review10.1145/3673660.365504552:1(1-2)Online publication date: 13-Jun-2024
  • (2024)Approximations to Study the Impact of the Service Discipline in Systems with RedundancyAbstracts of the 2024 ACM SIGMETRICS/IFIP PERFORMANCE Joint International Conference on Measurement and Modeling of Computer Systems10.1145/3652963.3655045(1-2)Online publication date: 10-Jun-2024
  • (2024)Approximations to Study the Impact of the Service Discipline in Systems with RedundancyProceedings of the ACM on Measurement and Analysis of Computing Systems10.1145/36390408:1(1-33)Online publication date: 21-Feb-2024
  • Show More Cited By

Recommendations

Comments

Information & Contributors

Information

Published In

cover image Queueing Systems: Theory and Applications
Queueing Systems: Theory and Applications  Volume 98, Issue 1-2
Jun 2021
202 pages

Publisher

J. C. Baltzer AG, Science Publishers

United States

Publication History

Published: 01 June 2021
Accepted: 18 January 2021
Revision received: 17 December 2020
Received: 19 June 2020

Author Tags

  1. Large-scale service systems
  2. Steady-state
  3. Asymptotic independence
  4. Multi-component jobs
  5. Redundancy
  6. Replication
  7. Cancel on start
  8. Cancel on completion
  9. Load distribution and balancing

Author Tags

  1. 90B15
  2. 60K25

Qualifiers

  • Research-article

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)0
  • Downloads (Last 6 weeks)0
Reflects downloads up to 30 Aug 2024

Other Metrics

Citations

Cited By

View all
  • (2024)Approximations to Study the Impact of the Service Discipline in Systems with RedundancyACM SIGMETRICS Performance Evaluation Review10.1145/3673660.365504552:1(1-2)Online publication date: 13-Jun-2024
  • (2024)Approximations to Study the Impact of the Service Discipline in Systems with RedundancyAbstracts of the 2024 ACM SIGMETRICS/IFIP PERFORMANCE Joint International Conference on Measurement and Modeling of Computer Systems10.1145/3652963.3655045(1-2)Online publication date: 10-Jun-2024
  • (2024)Approximations to Study the Impact of the Service Discipline in Systems with RedundancyProceedings of the ACM on Measurement and Analysis of Computing Systems10.1145/36390408:1(1-33)Online publication date: 21-Feb-2024
  • (2023)Join-Up-To(m): improved hyperscalable load balancingQueueing Systems: Theory and Applications10.1007/s11134-023-09897-5105:3-4(291-316)Online publication date: 1-Dec-2023

View Options

View options

Get Access

Login options

Media

Figures

Other

Tables

Share

Share

Share this Publication link

Share on social media