The multicore revolution is underway, bringing new chips introducing more complex memory architectures. Classical algorithms must be revisited in order to take the hierarchical memory layout into account. In this paper, we aim at designing cache-aware algorithms that minimize the number of cache misses paid during the execution of the matrix product kernel on a multicore processor. We analytically show how to achieve the best possible tradeoff between shared and distributed caches. We implement and evaluate several algorithms on two multicore platforms, one equipped with one Xeon quadcore, and the second one enriched with a GPU. It turns out that the impact of cache misses is very different across both platforms, and we identify what are th...
Cette thèse s intéresse aux algorithmes adaptés aux architectures mémoire hiérarchiques, rencontrées...
L'objectif de cette thèse est d'offrir des outils d'aide à la certification aéronautique de processe...
One of the challenges to achieving good performance on multicore architectures is the effective util...
The multicore revolution is underway, bringing new chips introducing more complex memory architectur...
The multicore revolution is underway. Classical algorithms have to be revisited in order to take hie...
nombre de pages: 25The multicore revolution is underway, bringing new chips introducing more complex...
International audienceThe multicore revolution is underway. Classical algorithms must be revisited i...
The multicore revolution is underway. Classi-cal algorithms have to be revisited in order to take hi...
Reordering instructions and data layout can bring significant performance improvement for memory bou...
The presence of shared caches in current multicore processors may generate a lot of performance vari...
The presence of shared caches in current multicore processors may generate a lot of performance vari...
Many current computer designs employ caches and a hierarchical memory architecture. The speed of a c...
Le concept de processeur multicœurs constitue le facteur dominant pour offrir des hautes performance...
Cette thèse s’intéresse aux algorithmes adaptés aux architectures mémoire hiérarchiques, rencontrées...
Chip multiprocessors (CMPs) have become virtually ubiquitous due to the increasing impact of power a...
Cette thèse s intéresse aux algorithmes adaptés aux architectures mémoire hiérarchiques, rencontrées...
L'objectif de cette thèse est d'offrir des outils d'aide à la certification aéronautique de processe...
One of the challenges to achieving good performance on multicore architectures is the effective util...
The multicore revolution is underway, bringing new chips introducing more complex memory architectur...
The multicore revolution is underway. Classical algorithms have to be revisited in order to take hie...
nombre de pages: 25The multicore revolution is underway, bringing new chips introducing more complex...
International audienceThe multicore revolution is underway. Classical algorithms must be revisited i...
The multicore revolution is underway. Classi-cal algorithms have to be revisited in order to take hi...
Reordering instructions and data layout can bring significant performance improvement for memory bou...
The presence of shared caches in current multicore processors may generate a lot of performance vari...
The presence of shared caches in current multicore processors may generate a lot of performance vari...
Many current computer designs employ caches and a hierarchical memory architecture. The speed of a c...
Le concept de processeur multicœurs constitue le facteur dominant pour offrir des hautes performance...
Cette thèse s’intéresse aux algorithmes adaptés aux architectures mémoire hiérarchiques, rencontrées...
Chip multiprocessors (CMPs) have become virtually ubiquitous due to the increasing impact of power a...
Cette thèse s intéresse aux algorithmes adaptés aux architectures mémoire hiérarchiques, rencontrées...
L'objectif de cette thèse est d'offrir des outils d'aide à la certification aéronautique de processe...
One of the challenges to achieving good performance on multicore architectures is the effective util...