The Gittins Policy in the M/G/1 Queue

  title={The Gittins Policy in the M/G/1 Queue},
  author={Ziv Scully and Mor Harchol-Balter},
  journal={2021 19th International Symposium on Modeling and Optimization in Mobile, Ad hoc, and Wireless Networks (WiOpt)},
  • Ziv ScullyMor Harchol-Balter
  • Published 18 October 2021
  • Computer Science
  • 2021 19th International Symposium on Modeling and Optimization in Mobile, Ad hoc, and Wireless Networks (WiOpt)
The Gittins policy is a highly general scheduling policy that minimizes a wide variety of mean holding cost metrics in the M/G/1 queue. Perhaps most famously, Gittins minimizes mean response time in the M/G/1 when jobs’ service times are unknown to the scheduler. Gittins also minimizes weighted versions of mean response time. For example, the well-known "cμ rule", which minimizes class-weighted mean response time in the multiclass M/M/1, is a special case of Gittins.However, despite the… 

Tables from this paper

Uniform Bounds for Scheduling with Job Size Estimates

A simple variant of SRPT is provided, which is the first result showing this type of convergence for M/G/1 scheduling, and it is proved its approximation ratio approaches 1 uniformly as α and β approach 1.

Optimal Scheduling in the Multiserver-job Model under Heavy Traffic

Multiserver-job systems, where jobs require concurrent service at many servers, occur widely in practice. Essentially all of the theoretical work on multiserver-job systems focuses on maximizing

The most common queueing theory questions asked by computer systems practitioners

This document examines five performance questions which are repeatedly asked by practitioners in industry: (i) My system utilization is very low, so why are job delays so high? (ii) What should I do

Minimizing the mean slowdown in a single-server queue

We consider the optimal scheduling problem in a single-server queue. Let Si and Ti denote the service time and the delay (a.k.a. sojourn time or response time) of job i . Service times are assumed to



The Gittins Policy is Nearly Optimal in the M/G/k under Extremely General Conditions

This work gives the first general analysis of Gittins in the M/G/k and shows that under extremely general conditions, Gittin's mean response time is at most itsmean response time in theM/g/1 plus an $O(łog(1/(1 - ρ)$ additive term, where ρ is the system load.

On the Gittins index in the M/G/1 queue

It is shown that in fact, Foreground–Background and First-Come-First-Served are optimal if and only if the service time distribution is of type Decreasing Hazard Rate and New Better than Used in Expectation, respectively.

Minimizing slowdown in heterogeneous size-aware dispatching systems

The main contribution of this paper is to show the optimality of SPTP with respect to slowdown in a single server queue under Poisson arrivals, and to derive the so-called size-aware value functions for M/G/1-FIFO/LifO/SPTP with general holding costs to derive efficient dispatching policies so as to minimize the mean slowdown in an heterogeneous server system.

Optimal Scheduling and Exact Response Time Analysis for Multistage Jobs

An optimal algorithm for scheduling multistage jobs in an M/G/1 queue is given and an exact response time analysis of the algorithm is given.

The Relation between Customer and Time Averages in Queues

In this note it is shown that Stidham's proof applies directly to the more general case of H = λG, provided λ and G are finite and a simple technical assumption is satisfied.


It turns out that the optimal policy always belongs to the family of multilevel processor sharing disciplines.

Characterizing Policies with Optimal Response Time Tails under Heavy-Tailed Job Sizes

The recently introduced SOAP framework is used to derive sufficient conditions on the form of prioritization used by a scheduling policy that ensure the policy is tail-optimal, and this work derives a general upper bound for fractional moments of M/G/1 busy periods.

Scheduling for Minimum Total Loss Using Service Time Distributions

The major results of the paper are the definition of a new scheduling rule based on the known service time distributions, and the proof that expected total loss is always minimized by using this new rule.

The achievable region method in the optimal control of queueing systems; formulations, bounds and policies

The central idea is to characterize the region of achievable performance in a stochastic control problem, i.e., find linear or nonlinear constraints on the performance vectors that all policies satisfy, to find heuristic nearly optimal policies.

Tax problems in the undiscounted case

The aim of this paper is to evaluate the performance of the optimal policy (the Gittins index policy) for open tax problems of the type considered by Klimov in the undiscounted limit. In this limit,