English

Achievable Stability in Redundancy Systems

Probability 2020-08-11 v1 Performance

Abstract

We consider a system with NN parallel servers where incoming jobs are immediately replicated to, say, dd servers. Each of the NN servers has its own queue and follows a FCFS discipline. As soon as the first job replica is completed, the remaining replicas are abandoned. We investigate the achievable stability region for a quite general workload model with different job types and heterogeneous servers, reflecting job-server affinity relations which may arise from data locality issues and soft compatibility constraints. Under the assumption that job types are known beforehand we show for New-Better-than-Used (NBU) distributed speed variations that no replication (d=1)(d=1) gives a strictly larger stability region than replication (d>1)(d>1). Strikingly, this does not depend on the underlying distribution of the intrinsic job sizes, but observing the job types is essential for this statement to hold. In case of non-observable job types we show that for New-Worse-than-Used (NWU) distributed speed variations full replication (d=Nd=N) gives a larger stability region than no replication (d=1)(d=1).

Keywords

Cite

@article{arxiv.2008.03478,
  title  = {Achievable Stability in Redundancy Systems},
  author = {Youri Raaijmakers and Sem Borst},
  journal= {arXiv preprint arXiv:2008.03478},
  year   = {2020}
}
R2 v1 2026-06-23T17:43:12.033Z