TY - GEN
T1 - Intralayer communication for tree-Based overlay networks
AU - Hilbrich, Tobias
AU - Protze, Joachim
AU - De Supinski, Bronis R.
AU - Schulz, Martin
AU - M̈uller, Matthias S.
AU - Nagel, Wolfgang E.
PY - 2013
Y1 - 2013
N2 - While various HPC tools use Tree-Based Overlay Networks (TBONs) to increase their scalability, some use cases do not map well to a tree-based hierarchy. We provide the concept of intralayer communication to improve this situation, where nodes in a specific hierarchy layer may exchange messages directly with each other. This concept targets data preprocessing that allows tool developers to avoid load imbalances in higher hierarchy levels. We implement intralayer communication within the Generic Tools Infrastructure (GTI) that provides TBON services, as well as a high-level abstraction to ease the creation of scalable runtime tools. An extension of GTI's abstractions allows simple and efficient use of intralayer communication. We demonstrate this capability with a runtime message matching tool for MPI's point-to-point communication, which we evaluate in an application study with up to 16,384 processes. Low overheads for two benchmark suites show the applicability of our approach, while a stress test demonstrates close to constant overheads across scales. The stress test measurements demonstrate that intralayer communication reduces application slowdown by two orders of magnitude at 2,048 processes, compared to a previous TBONbased implementation.
AB - While various HPC tools use Tree-Based Overlay Networks (TBONs) to increase their scalability, some use cases do not map well to a tree-based hierarchy. We provide the concept of intralayer communication to improve this situation, where nodes in a specific hierarchy layer may exchange messages directly with each other. This concept targets data preprocessing that allows tool developers to avoid load imbalances in higher hierarchy levels. We implement intralayer communication within the Generic Tools Infrastructure (GTI) that provides TBON services, as well as a high-level abstraction to ease the creation of scalable runtime tools. An extension of GTI's abstractions allows simple and efficient use of intralayer communication. We demonstrate this capability with a runtime message matching tool for MPI's point-to-point communication, which we evaluate in an application study with up to 16,384 processes. Low overheads for two benchmark suites show the applicability of our approach, while a stress test demonstrates close to constant overheads across scales. The stress test measurements demonstrate that intralayer communication reduces application slowdown by two orders of magnitude at 2,048 processes, compared to a previous TBONbased implementation.
UR - http://www.scopus.com/inward/record.url?scp=84893212015&partnerID=8YFLogxK
U2 - 10.1109/ICPP.2013.118
DO - 10.1109/ICPP.2013.118
M3 - Conference contribution
AN - SCOPUS:84893212015
SN - 9780769551173
T3 - Proceedings of the International Conference on Parallel Processing
SP - 995
EP - 1003
BT - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 42nd Annual International Conference on Parallel Processing, ICPP 2013
Y2 - 1 October 2013 through 4 October 2013
ER -