Sites Inria

Version française

Inria & HPC

Simgrid

An infrastructure simulator for better results in less time.

SimGrid is software that enables the study of large-scale distributed systems and the software that can be deployed on them. It can be used to assess algorithms, help with the development of distributed applications, determine execution times on all types of environment and benchmark the performances of material and software platforms in an extremely accurate manner.

Simgrid aims to address the following issues:

  • Inability to correctly predict execution times of complex applications (parallel, distributed, workflows) when there are time or financial constraints to respect,
  • Scaling problems of distributed material and software architectures,
  • Bugs that are difficult to solve or identify for parallel applications in a distributed environment that is complicated to manage

With Simgrid, users can:

  • Obtain better predictions of resolution times in deadline environments,
  • Achieve a better exploitation of computation infrastructure resources
  • Have a network infrastructure scaling capacity for extremely efficient and secure distributed applications
  • Drastically reduce bug correction times

Its network, CPU or hard drive models are compatible with the following environments: (Data)Grids, P2P, Clouds, Clusters and HPC, thereby enabling multi-field studies. It can be used not only for the simulation of algorithms and prototypes of applications but also to emulate MPI applications through the virtualisation of their communications or to formally verify the applications and algorithms capable of being executed using the framework.

Research team

The ROMA rsearch team aims at designing models, algorithms, and scheduling strategies to optimize the execution of scientific applications on High-Performance Computing platforms. More specifically, ROMA is interested in obtaining the “best” possible performance from the point of view of the user (e.g., application execution time) while using ressources as efficiently as possible (e.g., low energy consumption). The work performed by ROMA ranges from theoretical studies to the development of software used daily in the academic and industrial worlds.

More

Top