ABOUT US

RESEARCH

RESEARCHERS

REPORTS

SOFTWARE

FACILITIES

EMAIL SERVICES

WIKIS

AuthorDan Ibanez, Ian Dunn, Mark S Shephard
TitleHybrid MPI-thread Parallelization of Adaptive Mesh Operations
Year2014
JournalParallel Computing
Volume52
Pages133-143
PublisherElsevier
AbstractMany of the world's leading supercomputer architectures are a hybrid of shared memory and network-distributed memory. Such an architecture lends itself to a hybrid MPI-thread programming model. We first present an implementation of inter-thread message passing based on the MPI and pthread libraries. In addition, an efficient implementation of termination detection for communication rounds is presented called phased message passing. This interface is then used to implement parallel operations for adaptive unstructured meshes, and the performance of resulting applications is compared to pure MPI operation. We also present new workflows enabled by the ability to vary the number of threads during runtime.
PDF File Download
DOI Link10.1016/j.parco.2016.01.003