TY - JOUR
T1 - Tiling on systems with communication/computation overlap
AU - Calland, Pierre Yves
AU - Dongarra, Jack
AU - Robert, Yves
PY - 1999/3
Y1 - 1999/3
N2 - In the framework of fully permutable loops, thing is a compiler technique (also known as 'loop blocking') that has been extensively studied as a source-to-source program transformation. Little work has been devoted to the mapping and scheduling of the tiles on to physical parallel processors. We present several new results in the context of limited computational resources and assuming communication-computation overlap. In particular, under some reasonable assumptions, we derive the optimal mapping and scheduling of tiles to physical processors.
AB - In the framework of fully permutable loops, thing is a compiler technique (also known as 'loop blocking') that has been extensively studied as a source-to-source program transformation. Little work has been devoted to the mapping and scheduling of the tiles on to physical parallel processors. We present several new results in the context of limited computational resources and assuming communication-computation overlap. In particular, under some reasonable assumptions, we derive the optimal mapping and scheduling of tiles to physical processors.
U2 - 10.1002/(SICI)1096-9128(199903)11:3<139::AID-CPE370>3.0.CO;2-X
DO - 10.1002/(SICI)1096-9128(199903)11:3<139::AID-CPE370>3.0.CO;2-X
M3 - Article
VL - 11
SP - 139
EP - 153
JO - Concurrency: Practice and Experience
JF - Concurrency: Practice and Experience
IS - 3
ER -