Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

[Licence| Download | New Version Template] aeez_v1_0.tar.gz(25306 Kbytes)
Manuscript Title: MonALISA: An Agent Based, Dynamic Service System to Monitor, Control and Optimize Distributed Systems
Authors: I. Legrand, H. Newman, R. Voicu, C. Cirstoiu, C. Grigoras, C. Dobre, A. Muraru, A. Costan, M. Dediu, C. Stratan
Program title: MonALISA
Catalogue identifier: AEEZ_v1_0
Distribution format: tar.gz
Journal reference: Comput. Phys. Commun. 180(2009)2472
Programming language: Java, additional APIs available in Java, C, C++, perl and python.
Computer: Computing Clusters, Network Devices, Storage Systems, Large scale data intensive applications.
Operating system: The MonALISA service is mainly used in Linux, the MonALISA client runs on all major platforms ( Windows, Linux, Solaris, MacOS ).
Has the code been vectorised or parallelized?: It is a multithreaded application. It will efficiently use all the available processors.
RAM: for the MonALISA service the minimum required memory is 64 MB; if the JVM is started allocating more memory this will be used for internal caching. The MonALISA client requires typically 256-512 MB of memory.
Keywords: distributed computing systems, monitoring, control, optimized dynamic routing, high performance data transfers.
PACS: 07.05.Bx, 07.05.Dz, 07.05.Rm.
Classification: 6.5.

External routines: requires Java: JRE or JDK to run. These external packages are used (they are included in the distribution): JINI, JFreeChart, PostgreSQL (optional).

Nature of problem:
To monitor and control distributed computing clusters and grids, the network infrastructure, the storage systems, and the applications used on such facilities. The monitoring information gathered is used for developing the required higher level services, the components that provide decision support and some degree of automated decisions and for maintaining and optimizing workflow in large scale distributed systems.

Solution method:
The MonALISA framework is designed as an ensemble of autonomous self-describing agent-based subsystems which are registered as dynamic services. These services are able to collaborate and cooperate in performing a wide range of distributed information-gathering and processing tasks.

Running time:
MonALISA services are designed to run continuously to collect monitoring data and to trigger alarms or to take automatic actions in case it is necessary.

References:
[1] http://monalisa.caltech.edu