TY - GEN
T1 - Overcoming scalability challenges for tool daemon launching
AU - Ahn, Dong H.
AU - Arnold, Dorian C.
AU - De Supinski, Bronis R.
AU - Lee, Gregory L.
AU - Miller, Barton P.
AU - Schulz, Martin
PY - 2008
Y1 - 2008
N2 - Many tools that target parallel and distributed environments must co-locate a set of daemons with the distributed processes of the target application. However, efficient and portable deployment of these daemons on large scale systems is an unsolved problem. We overcome this gap with LaunchMON, a scalable, robust, portable, secure, and general purpose infrastructure for launching tool daemons. Its API allows tool builders to identify all processes of a target job, launch daemons on the relevant nodes and control daemon interaction. Our results show that LaunchMON scales to very large daemon counts and substantially enhances performance over existing ad hoc mechanisms.
AB - Many tools that target parallel and distributed environments must co-locate a set of daemons with the distributed processes of the target application. However, efficient and portable deployment of these daemons on large scale systems is an unsolved problem. We overcome this gap with LaunchMON, a scalable, robust, portable, secure, and general purpose infrastructure for launching tool daemons. Its API allows tool builders to identify all processes of a target job, launch daemons on the relevant nodes and control daemon interaction. Our results show that LaunchMON scales to very large daemon counts and substantially enhances performance over existing ad hoc mechanisms.
UR - http://www.scopus.com/inward/record.url?scp=55849091769&partnerID=8YFLogxK
U2 - 10.1109/ICPP.2008.63
DO - 10.1109/ICPP.2008.63
M3 - Conference contribution
AN - SCOPUS:55849091769
SN - 9780769533742
T3 - Proceedings of the International Conference on Parallel Processing
SP - 578
EP - 585
BT - Proceedings - 37th International Conference on Parallel Processing, ICPP 2008
T2 - 37th International Conference on Parallel Processing, ICPP 2008
Y2 - 9 September 2008 through 12 September 2008
ER -