Supermon: a high-speed cluster monitoring system

Sottile, Matthew; Minnich, Ronald G.

doi:10.1109/clustr.2002.1137727

Cited by 79 publications

(46 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…TAUg has shown low overheads up to 512 processors. TAUoverSupermon takes a different strategy from TAUg, using the Supermon cluster monitoring framework [15] as the data aggregation mechanism (referred to in their paper as the "transport"). It too, shows low overhead up to 512 processors.…”

Section: B Types Of Existing Performance Analysis Toolsmentioning

confidence: 99%

Continuous performance monitoring for large-scale parallel applications

Dooley

Lee

Kalé

2009

2009 International Conference on High Performance Computing (HiPC)

View full text Add to dashboard Cite

Abstract-Traditional performance analysis techniques are performed after a parallel program has completed. In this paper, we describe an online method for continuously monitoring the performance of a parallel program, specifically the fraction of the time spent in various activities as the program executes. Our implementation of both a visualization client and the parallel performance framework that gathers utilization data are described. The data gathering uses a scalable and asynchronous reduction with an appropriate lossless compressed data format. The overheads in the initial system are low, even when run on thousands of processors. The data gathering occurs in an out-ofband communication mechanism, interleaving itself transparently with the execution of the parallel application by leveraging a message-driven runtime system.

show abstract

Section: B Types Of Existing Performance Analysis Toolsmentioning

confidence: 99%

Continuous performance monitoring for large-scale parallel applications

Dooley

Lee

Kalé

2009

2009 International Conference on High Performance Computing (HiPC)

View full text Add to dashboard Cite

show abstract

“…It relies on symbolic expressions at all levels for communication between components in order to reduce node perturbation when parsing messages [27]. A single mon process runs on all the compute nodes, they parse the symbolic expressions from the kernel module and make that information available over a TCP port to clients and the szlpermon data aggregator.…”

Section: Supermonmentioning

confidence: 99%

“…Querying each node in the cluster is important to achieve the third goal of scalability since this time will only increase as cluster sizes become larger. To accomplish this goal, the Fountain daemons are arranged in a tree topology as described in act as a wrapper around a more specialized monitoring component like Supermon [27], and present its data using the SSS interface.…”

Section: Query Performancementioning

confidence: 99%

The Node Monitoring Component of a Scalable Systems Software Environment

Miller

2006

View full text Add to dashboard Cite

“…Recently, several projects have begun to address this problem by integrating hierarchical communication structures with online aggregation mechanisms, like MRNet [2] or Supermon [3], into their tools. On the debugger side, HP's Ladebug [4] relies on a tree of debug daemons to control large…”

Section: Introductionmentioning

confidence: 99%

Lessons learned at 208K: Towards debugging millions of cores

L¹,

Ahn²,

Arnold³

et al. 2008

2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

Abstract-Petascale systems will present several new challenges to performance and correctness tools. Such machines may contain millions of cores, requiring that tools use scalable data structures and analysis algorithms to collect and to process application data. In addition, at such scales, each tool itself will become a large parallel application -already, debugging the full BlueGene/L (BG/L) installation at the Lawrence Livermore National Laboratory requires employing 1664 tool daemons. To reach such sizes and beyond, tools must use a scalable communication infrastructure and manage their own tool processes efficiently. Some system resources, such as the file system, may also become tool bottlenecks.In this paper, we present challenges to petascale tool development, using the Stack Trace Analysis Tool (STAT) as a case study. STAT is a lightweight tool that gathers and merges stack traces from a parallel application to identify process equivalence classes. We use results gathered at thousands of tasks on an Infiniband cluster and results up to 208K processes on BG/L to identify current scalability issues as well as challenges that will be faced at the petascale. We then present implemented solutions to these challenges and show the resulting performance improvements. We also discuss future plans to meet the debugging demands of petascale machines.

show abstract

Supermon: a high-speed cluster monitoring system

Cited by 79 publications

References 3 publications

Continuous performance monitoring for large-scale parallel applications

Continuous performance monitoring for large-scale parallel applications

The Node Monitoring Component of a Scalable Systems Software Environment

Lessons learned at 208K: Towards debugging millions of cores

Contact Info

Product

Resources

About