A low impact flow control implementation for offload communication interfaces

Authors:
Brian W. Barrett;Ron Brightwell;Keith D. Underwood
Affiliations:
Sandia National Laboratories, Albuquerque, NM;Sandia National Laboratories, Albuquerque, NM;Intel Corporation, Hillsboro, OR
Venue:
EuroMPI'12 Proceedings of the 19th European conference on Recent Advances in the Message Passing Interface
Year:
2012

Citing 6
Cited 1

The Quadrics Network: High-Performance Clustering Technology

IEEE Micro
Portals 3.0: Protocol Building Blocks for Low Overhead Communication

IPDPS '02 Proceedings of the 16th International Parallel and Distributed Processing Symposium
A Scalable Flow Control Algorithm for the Fast Messages Communication Library

CANPC '99 Proceedings of the Third International Workshop on Network-Based Parallel Computing: Communication, Architecture, and Applications
GASNet Specification, v1.1

GASNet Specification, v1.1
Scaling MPI to short-memory MPPs such as BG/L

Proceedings of the 20th annual international conference on Supercomputing
Enabling Flexible Collective Communication Offload with Triggered Operations

HOTI '11 Proceedings of the 2011 IEEE 19th Annual Symposium on High Performance Interconnects

Design, implementation, and performance evaluation of MPI 3.0 on portals 4.0

Proceedings of the 20th European MPI Users' Group Meeting

Quantified Score

Hi-index	0.00

Visualization

Abstract

Message passing paradigms provide for many to one messaging patterns that result in receive side resource exhaustion. Traditionally, MPI implementations layered over the Portals network programming interface provided a large default unexpected receive buffer space, the user was expected to configure the buffer size to the application demand, and the application was aborted when the buffer space was overrun. The Portals 4 design provides a set of primitives for implementing scalable resource exhaustion recovery without negatively impacting normal operation. A resource exhaustion recovery protocol for MPI implementations is presented, as well as performance results for an Open MPI implementation of the protocol.