Library Miniaturization Using Static and Dynamic Information

  • Authors:
  • Giuliano Antoniol;Massimiliano Di Penta

  • Affiliations:
  • -;-

  • Venue:
  • ICSM '03 Proceedings of the International Conference on Software Maintenance
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Moving to smaller libraries can be considered as a relevanttask when porting software systems to limited-resourcedevices (e.g., hand-helds). Library miniaturization will beparticularly effective if based on both dynamic (keeping intoaccount dependencies exploited during application executionin a given user profile) and static (keeping into accountall possible dependencies) information.This paper presents a distributed software architecture,based on web services, to collect dynamic information atrun-time, and an approach for miniaturization of libraries,exploiting both dynamic and static information with the aimof reducing the memory requirements of executables.New, smaller libraries are identified via hierarchicalclustering and genetic algorithms; clustering produces afirst initial solution, then optimized by multi-objective geneticalgorithms.The approach has been applied to medium size opensource software systems such as Samba and MySQL, allowingto effectively produce smaller, loosely coupled libraries,and to reduce the memory requirements of each application.