An heuristic model for analysis of memory use under static partition allocation strategies

  • Authors:
  • Trân Quôc Tê

  • Affiliations:
  • -

  • Venue:
  • SIGMETRICS '76 Proceedings of the 1976 ACM SIGMETRICS conference on Computer performance modeling measurement and evaluation
  • Year:
  • 1976

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper examines minimum memory wastage strategies for partitioning and allocating main memory to a stream of jobs, in a context where each job resides throughout its execution in a frame to which it is assigned. In contrast to usual approaches, the present one considers core memory as the main limiting resource, and assumes that job competition for other resources can be neglected. From job mix characteristics, a condition for a given memory partition to avoid system saturation is postulated. Then, among such partitions avoiding saturation, we are interested in 1) those requiring the least amount of space, 2) those leading to the lowest core fragmentation. Concerning the first problem, we give a simple method for generating the smallest partition, and show that “exact fit” allocation should be used. As for the second problem, if a partition is given so that exact fit meets the overload constraint then this algorithm is optimal; otherwise, best fit would be the best algorithm.