Copyright Notice:

The documents distributed by this server have been provided by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a noncommercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.

Publications of SPCL

T. Hoefler, A. Lumsdaine:

 Optimizing non-blocking Collective Operations for InfiniBand

(. Vol , Nr. , In Proceedings of the 22nd IEEE International Parallel & Distributed Processing Symposium, CAC'08 Workshop, presented in Miami, FL, pages , , ISSN: 1530-2075, ISBN: 978-1-4244-1694-3, Apr. 2008)


Non-blocking collective operations have recently been shown to be a promising complementary approach for overlapping communication and computation in parallel applications. However, in order to maximize the performance and usability of these operations it is important that they progress concurrently with the application without introducing CPU overhead and without requiring explicit user intervention. While studying non-blocking collective operations in the context of our portable library (libNBC), we found that most MPI implementations do not sufficienctly support overlap over the InfiniBand network. To address this issue, we developed a low-level communication layer for libNBC based on the Open Fabrics InfiniBand verbs API. With this layer we are able to achieve high degrees of overlap without the need to explicitly progress the communication operations. We show that the communication overhead of parallel application kernels can be reduced up to 92% while not requiring user intervention to make progress.


download article:
download slides:


  author={Torsten Hoefler and Andrew Lumsdaine},
  title={{Optimizing non-blocking Collective Operations for InfiniBand}},
  booktitle={Proceedings of the 22nd IEEE International Parallel \& Distributed Processing Symposium, CAC'08 Workshop},
  location={Miami, FL},