Speed Scaling with Multiple Servers Under A Sum Power Constraint

TL;DR

This paper studies scheduling multiple servers under a total power limit to minimize combined flow time and energy, introducing algorithms with proven competitive ratios for jobs with diminishing returns from parallelization.

Contribution

It presents simple algorithms for multi-server scheduling with concave speedup functions and analyzes their competitive ratios in both static and dynamic job arrival scenarios.

Findings

EQUI algorithm is $ig(2-rac{1}{ ext{alpha}}ig) rac{2}{1-rac{1}{ ext{alpha}}}$-competitive for static jobs.

LCFS-based algorithm has a constant competitive ratio for dynamic job arrivals.

The approach generalizes flow time minimization to jobs with sub-linear speedup functions.

Abstract

The problem of scheduling jobs and choosing their respective speeds with multiple servers under a sum power constraint to minimize the flow time + energy is considered. This problem is a generalization of the flow time minimization problem with multiple unit-speed servers, when jobs can be parallelized, however, with a sub-linear, concave speedup function $k^{1/\alpha}, \alpha>1$ when allocated $k$ servers, i.e., jobs experience diminishing returns from being allocated additional servers. When all jobs are available at time $0$, we show that a very simple algorithm EQUI, that processes all available jobs at the same speed is $\left(2-\frac{1}{\alpha}\right) \frac{2}{\left(1-\left(\frac{1}{\alpha}\right)\right)}$-competitive, while in the general case, when jobs arrive over time, an LCFS based algorithm is shown to have a constant (dependent only on $\alpha$) competitive ratio.