Speaker
I. Legrand
(CALTECH)
Description
The MonALISA (MONitoring Agents in A Large Integrated Services Architecture) system
is a scalable Dynamic Distributed Services Architecture which is based on the mobile
code paradigm.
An essential part of managing a global system, like the Grids, is a monitoring system
that is able to monitor and track the many site facilities, networks, and all the
task in progress, in real time. MonALISA is designed to easily integrate existing
monitoring tools and procedures and to provide this information in a dynamic, self
describing way to any other services or clients.
The monitoring information gathered is essential for developing higher level
services that provide decision support, and eventually some degree of automated
decisions, to help maintain and optimize workflow through the Grid.
MonALISA is an ensemble of autonomous multi-threaded, agent-based subsystems which
are registered as dynamic services and are able to collaborate and cooperate in
performing a wide range of monitoring, data processing and control tasks in large
scale distributed applications. We also present the development of specialized higher
level services, implemented as distributed mobile agents in the MonALISA framework to
control and globally optimize tasks as grid scheduling, real-time data streaming or
effective file replication.I
The system is currently used to monitor several large scale systems and provides
detailed information for computing nodes, LAN and WAN network components, job
execution and applications specific parameters. This distributed system proved to be
reliable, able to correctly handle connectivity problems and is running around the
clock on more than 120 sites.
Primary authors
C Grigoras
(Polytechnic University Bucharest)
C. Cirstoiu
(Polytechnic University Bucharest)
C. Dobre
(Polytechnic University Bucharest)
H. Newman
(Caltech)
I. Legrand
(CALTECH)
M. Toarta
(Polytechnic University Bucharest)
R. Voicu
(CERN)
