No. 131 (00308)

Family name : Legrand 
Given name : Iosif 
Affiliation : California Institute of Technology 
Abbreviation : CATLECH 
E-mail address : Iosif.Legrand@cern.ch 
Title : A Processes Oriented, Discrete Event Simulation Framework for
        Modelling and Design of Large Scale Distributed Systems
Authors : I.C.Legrand, H.B. Newman,  F. Lingen, C. Dobre , C. Stratan,
          K. Paschen 
Abstract : 
The design and optimization of   Computing Models for the future LHC
experiments, based on the GRID technologies, requires a realistic and
effective  modeling of  the  data  access  patterns,  the  data  flow
across  the  local and wide area networks, and the scheduling and
workload presented  by  hundreds  of  data intensive  jobs  running
concurrently on large scale distributed systems. 

This paper discusses the development of the MONARC (MOdels of Networked
Analysis at Regional Centers) simulation framework, as a design and
modelling tool for large scale distributed systems.  A  process-oriented
approach  for  discrete  event simulation  is well suited  for
describing  concurrent  running programs, as well as all the stochastic
arrival patterns that characterize how such systems are used.  The
simulation engine is based on Threaded Objects, (or Active Objects)
which offer great flexibility in simulating the complex behavior of
distributed data processing programs.  The engine provides an
appropriate scheduling mechanism for the Active objects with support for
interrupts.  This approach offers a natural way of describing complex
running programs that are data dependent and which concurrently compete
for shared resources as well as large numbers of concurrent data
transfers on shared resources.  

The framework provides a complete set of basic components (processing
nodes, data servers, network components) together with dynamically
loadable decision units (scheduling or data replication modules) for
easily building complex Computing Models simulations.   Examples of
simulating complex data processing systems are presented and the way the
framework is used to compare different decision making algorithms or to
optimize the architecture.  

The evaluation of Computing Models for the LHC experiments involves
multiple iterations of monitoring,   modelling and understanding of
current systems.  The MonALISA (MONitoring Agents using a Large
Integrated Service Architecture) framework is used for monitoring and
understanding the performance of current prototypes, and input this
information into the simulation system.