Imagen de Portada
Código QR

Refactoring for Data Locality.

The article discusses refactoring software programs to improve execution speed with respect to caching and data locality. Topics include cache misses due to multiple intervening cache use, the computer code responsible for such “long distance” reusing of data, how the reuse distance can be shortened...

Descripción completa

Publicado en:Computer (00189162) Vol. 42; no. 2; pp. 62 - 72
Autores principales: Beyls, Kristof, D'Hollander, Erik H.
Formato: Artículo
Publicado: IEEE Feb2009
Materias:
Software engineering
Cache memory
Computer programming
Program transformation
Computer storage devices
Algorithms
Acceso en línea:Ver este registro en EBSCOhost
Sumario:The article discusses refactoring software programs to improve execution speed with respect to caching and data locality. Topics include cache misses due to multiple intervening cache use, the computer code responsible for such “long distance” reusing of data, how the reuse distance can be shortened, and using suggestions for locality optimizations (SLO) for program optimization. Case studies are discussed that involve the use of SLO to examine seven SPEC2000 benchmark programs. The authors regard SLO as having significant value in optimizing data-intensive applications.