Thinking Clearly About Performance, Part 1.

The article presents information on axiomatic methods for optimizing computer software performance. The usefulness of axiomatic reasoning for ensuring clear and testable improvements is discussed. It is noted that the majority of software performance improvements involve speed of response. Various e...

Descripción completa

Detalles Bibliográficos
Publicado en:Communications of the ACM Vol. 53; no. 9; pp. 55 - 61
Autor principal: MILLSAP, CARY
Formato: Artículo
Publicado: Association for Computing Machinery Sep2010
Materias:
Acceso en línea:Ver este registro en EBSCOhost
Descripción
Sumario:The article presents information on axiomatic methods for optimizing computer software performance. The usefulness of axiomatic reasoning for ensuring clear and testable improvements is discussed. It is noted that the majority of software performance improvements involve speed of response. Various examples of how a solution which seems to work well in theory can actually fail in practice are provided. Replacing simple averages with distribution percentages is recommended as a means of identifying problems more accurately.