Analysis and optimization of a demographic simulator for parallel environments

Description 

The aim of this project is the exploitation of the capacities offered by multicore architecture in order to improve the performance of the social demographic simulator, Yades (Yet Another DEmographic Simulator), created in inLab in collaboration with Lancaster University and Barcelona Supercomputing Center (BSC)..

To accomplish this objective, ROOT-Sim (ROme OpTimistic Simulator)) library has been used, taking profit of the benefits that a shared memory architecture offers and optimizing the performance of the simulator. This enhancement reduces intra-node communications and provides more advanced capabilities for working with highly extended architecture currently.

To achieve this goal, we initiate a collaboration with High Performance and Dependable Computing Systems Group (DIAG) of Sapienza University, in order to adapt our simulator with ROOT-Sim library, a parallel simulation general purpose library for discrete events (PDES - Parallel Discrete Event Simulation). This static library can be used with any model that uses ANSI-C standard.

With this project not only has been optimized the performance of Yades simulator in multicore architecture but also has contributed in the development of some parts of ROOT-Sim library and its test with a real application.

Duration of the project 
January, 2015 to July, 2015
Technology 
MPI (Message Passing Interface),
PTHREAD (POSIX Threads),
GitHub,
C,
XML,
C++,
Simulador YADES,
VMWare Workstation
Areas of expertise involved in the project 
Project Manager 
Participants 
Alessandro Pellegrini
Thesis, master's or degree's final projects 
Articles and Presentations 

Follow us on

Els nostres articles del bloc d'inLab FIB

         
         

inLab FIB incorporates esCert

Icona ESCERT

inLab is member of