Composability of parallel codes on heterogeneous architectures

Abstract : To face the ever demanding requirements in term of accuracy and speed of scientific simulations, the High Performance community is constantly increasing the demands in term of parallelism, adding thus tremendous value to parallel libraries strongly optimized for highly complex architectures.Enabling HPC applications to perform efficiently when invoking multiple parallel libraries simultaneously is a great challenge. Even if a uniform runtime system is used underneath, scheduling tasks or threads coming from dfferent libraries over the same set of hardware resources introduces many issues, such as resource oversubscription, undesirable cache ushes or memory bus contention.In this thesis, we present an extension of StarPU, a runtime system specifically designed for heterogeneous architectures, that allows multiple parallel codes to run concurrently with minimal interference. Such parallel codes run within scheduling contexts that provide confined executionenvironments which can be used to partition computing resources. Scheduling contexts can be dynamically resized to optimize the allocation of computing resources among concurrently running libraries. We introduced a hypervisor that automatically expands or shrinks contexts using feedback from the runtime system (e.g. resource utilization). We demonstrated the relevance of this approach by extending an existing generic sparse direct solver (qr mumps) to use these mechanisms and introduced a new decomposition method based on proportional mapping that is used to build the scheduling contexts. In order to cope with the very irregular behavior of the application, the hypervisor manages dynamically the allocation of resources. By means of the scheduling contexts and the hypervisor we improved the locality and thus the overall performance of the solver.
Document type :
Theses
Complete list of metadatas

Cited literature [55 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-01162975
Contributor : Abes Star <>
Submitted on : Thursday, June 11, 2015 - 5:47:05 PM
Last modification on : Wednesday, April 4, 2018 - 1:24:20 AM
Long-term archiving on : Tuesday, April 25, 2017 - 7:05:09 AM

File

HUGO_ANDRA_2014.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-01162975, version 1

Citation

Andra-Ecaterina Hugo. Composability of parallel codes on heterogeneous architectures. Computer science. Université de Bordeaux, 2014. English. ⟨NNT : 2014BORD0373⟩. ⟨tel-01162975⟩

Share

Metrics

Record views

400

Files downloads

624