A Lower Bound for Scheduling of Unit Jobs with Immediate Decision on Parallel Machines

  • Authors:
  • Tomáš Ebenlendr;Jiří Sgall

  • Affiliations:
  • Institute of Mathematics, AS CR, Praha 1, Czech Republic CZ-11567;Institute of Mathematics, AS CR, Praha 1, Czech Republic CZ-11567

  • Venue:
  • Approximation and Online Algorithms
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Consider scheduling of unit jobs with release times and deadlines on m identical machines with the objective to maximize the number of jobs completed before their deadlines. We prove a new lower bound for online algorithms with immediate decision. This means that the jobs arrive over time and the algorithm has to decide the schedule of each job immediately upon its release. Our lower bound tends to e/(e − 1) ≅ 1.58 for many machines, matching the performance of the best algorithm.