Balanced Fair Resource Sharing in Computer Clusters
arXiv:1604.06763 · doi:10.1016/j.peva.2017.08.006
Abstract
We represent a computer cluster as a multi-server queue with some arbitrary bipartite graph of compatibilities between jobs and servers. Each server processes its jobs sequentially in FCFS order. The service rate of a job at any given time is the sum of the service rates of all servers processing this job. We show that the corresponding queue is quasi-reversible and use this property to design a scheduling algorithm achieving balanced fair sharing of the service capacity.
References in corpus (1)
Cited by in corpus (9)
- On the stability of redundancy models
- Stochastic Non-Bipartite Matching Models and Order-Independent Loss Queues
- A token-based central queue with order-independent service rates
- Improving the performance of heterogeneous data centers through redundancy
- Poly-Symmetry in Processor-Sharing Systems
- Pass-and-Swap Queues
- Approximations to Study the Impact of the Service Discipline in Systems with Redundancy
- Dynamic Load Balancing with Tokens
- Distributed Server Allocation for Content Delivery Networks