Abstract

In many mission-critical applications such as police and homeland security-related information systems, automated monitoring of relevant information sources is essential. Such monitoring results in a large number of periodic queries, which can significantly increase the load on a server that hosts information services. If the execution of these queries is not carefully scheduled on the server, high peak load might occur, leading to degraded service quality. We investigate this query scheduling problem with the objective of minimizing the server’s peak load. We state an optimization-based formulation and show that this problem is NP-hard in the strong sense. Subsequently, several greedy heuristic approaches are developed and compared via a computational study with respect to solution quality and computational efficiency.

Keywords: Scheduling; Greedy heuristics; Periodic queries

Article Outline

1. Introduction

2. Related work

3. Problem modeling

3.1. Problem definition

3.2. A numerical example

3.3. The PM problem formulation

3.4. Lower bounds

3.4.1. Lower bound I

3.4.2. Lower bound II

3.4.3. Lower bound III

4. Greedy methods for solving the PM problem

4.1. Static greedy methods

Static greedy heuristics

4.2. A dynamic greedy heuristic

Dynamic greedy heuristic (DH)

4.3. A pairwise look-ahead greedy heuristic

A pairwise look-ahead greedy heuristic (PAIR)

4.4. Worst-case analysis of greedy methods

5. A computational study

5.1. PM problem instance generation

5.2. Solution quality comparison

5.2.1. Small PM instances (10 users)

5.2.2. Medium to large PM instances (50, 200, 1000 users)

5.3. Computational efficiency comparison

6. Conclusion and future research

Acknowledgements

Appendix A. Proof of Theorem 1

References