Efficient multithreaded context ID allocation in MPI

Authors:
James Dinan;David Goodell;William Gropp;Rajeev Thakur;Pavan Balaji
Affiliations:
Argonne National Laboratory;Argonne National Laboratory;University of Illinois at Urbana-Champaign;Argonne National Laboratory;Argonne National Laboratory
Venue:
EuroMPI'12 Proceedings of the 19th European conference on Recent Advances in the Message Passing Interface
Year:
2012

Citing 6
Cited 1

A multithreaded message passing interface (MPI) architecture: performance and program issues

Journal of Parallel and Distributed Computing
Optimizing threaded MPI execution on SMP clusters

ICS '01 Proceedings of the 15th international conference on Supercomputing
Thread-safety in an MPI implementation: Requirements and analysis

Parallel Computing
Scalability of communicators and groups in MPI

Proceedings of the 19th ACM International Symposium on High Performance Distributed Computing
Noncollective communicator creation in MPI

EuroMPI'11 Proceedings of the 18th European MPI Users' Group conference on Recent advances in the message passing interface
Load Balancing of Dynamical Nucleation Theory Monte Carlo Simulations through Resource Sharing Barriers

IPDPS '12 Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium

Dataflow coordination of data-parallel tasks via MPI 3.0

Proceedings of the 20th European MPI Users' Group Meeting

Quantified Score

Hi-index	0.00

Visualization

Abstract

An important aspect of support for multithreaded MPI executions is the management of communication context identifiers (IDs), which are used to associate MPI communication operations with a communicator. New communicator creation functionality in MPI 3.0 adds complexity to this core resource management problem. We present an efficient algorithm for multithreaded context ID allocation that builds on an existing production algorithm developed to support MPI 2.2. Through this work, we have discovered a subtle concurrency bug in the existing algorithm that can result in deadlock. We correct this bug and develop methods to overcome the performance impact of deadlock prevention. We evaluate the performance of the new algorithm and prove that it is free from deadlock.