In modern parallel architectures, memory accesses represent a commonbottleneck. Thus, optimizing the way applications access the memory is an important way to improve performance and energy consumption. Memory accesses are even more important with NUMAmachines, as the access time to data depends on its location inthe memory. Many efforts were made todevelop adaptive tools to improve memory accesses at the runtime by optimizingthe mapping of data and threads to NUMA nodes. However, theses tools are notable to change the memory access pattern of the original application,therefore a code written without considering memory performance mightnot benefit from them. Moreover, automatic mapping tools take time to convergetowards the best mapping, l...
La multiplication des processeurs et des coeurs dans les machines a conduit les architectes à renonc...
Miniaturization of electronic components has led to the introduction of complex electronic systems w...
The field of High Performance Computing (HPC) is characterized by the continuous evolution of comput...
Modern multicore systems are based on a Non-Uniform Memory Access (NUMA) design. In a NUMA system, c...
In modern High Performance Computing architectures, the memory subsystem is a common performance ...
Task-based models and runtimes are quite popular in the HPC community. Theyhelp to implement applica...
Parallel programs need to manage the time trade-off between synchronization and computation. A high ...
Next generation high performance computers will massively use virtualization as a way to share hardw...
The Kalray MPPA3 Coolidge many-core processor is one of the few off-the-shelf high-performance proce...
Current supercomputer architectures are subject to memory related issues. For instance we can observ...
Scientific applications have an increasing need of resources and many grand scientific challenges re...
Corresponding publication at IMPACT'16 (http://impact.gforge.inria.fr/impact2016)In this report, we ...
This thesis report is focused on studying data flow accelerators dedicated to image using mathematic...
Since several years, classical multiprocessor systems have evolved to multicores, which tightly inte...
At first, we were interested in TimSort, a sorting algorithm which was designed in 2002, at a time w...
La multiplication des processeurs et des coeurs dans les machines a conduit les architectes à renonc...
Miniaturization of electronic components has led to the introduction of complex electronic systems w...
The field of High Performance Computing (HPC) is characterized by the continuous evolution of comput...
Modern multicore systems are based on a Non-Uniform Memory Access (NUMA) design. In a NUMA system, c...
In modern High Performance Computing architectures, the memory subsystem is a common performance ...
Task-based models and runtimes are quite popular in the HPC community. Theyhelp to implement applica...
Parallel programs need to manage the time trade-off between synchronization and computation. A high ...
Next generation high performance computers will massively use virtualization as a way to share hardw...
The Kalray MPPA3 Coolidge many-core processor is one of the few off-the-shelf high-performance proce...
Current supercomputer architectures are subject to memory related issues. For instance we can observ...
Scientific applications have an increasing need of resources and many grand scientific challenges re...
Corresponding publication at IMPACT'16 (http://impact.gforge.inria.fr/impact2016)In this report, we ...
This thesis report is focused on studying data flow accelerators dedicated to image using mathematic...
Since several years, classical multiprocessor systems have evolved to multicores, which tightly inte...
At first, we were interested in TimSort, a sorting algorithm which was designed in 2002, at a time w...
La multiplication des processeurs et des coeurs dans les machines a conduit les architectes à renonc...
Miniaturization of electronic components has led to the introduction of complex electronic systems w...
The field of High Performance Computing (HPC) is characterized by the continuous evolution of comput...