Dynamic Malleability in MPI Applications

Dynamic Malleability in MPI Applications Kaoutar El Maghraoui Travis J. Desell Boleslaw K. Szymanski Carlos A. Varela Malleability enables a parallel application’s execution system to split or merge processes modifying the parallel application’s granularity. While process migration is widely used to adapt applications to dynamic execution environments, it is limited by the granularity of the application’s processes. Malleability empowers process migration by allowing the application’s processes to expand or shrink following the availability of resources. We have implemented malleability as an extension to the PCM (Process Checkpointing and Migration) library, a user-level library for iterative MPI applications. PCM is integrated with the Internet Operating System (IOS), a framework for middleware-driven dynamic application reconfiguration. Our approach requires minimal code modifications and enables transparent middleware-triggered reconfiguration. We present experimental results that demonstrate the usefulness of malleability. Department of Computer Science, Rensselaer Polytechnic Institute, Troy, NY cs-06-17

Dynamic Malleability in MPI Applications

Kaoutar El Maghraoui

Travis J. Desell

Boleslaw K. Szymanski

Carlos A. Varela

Malleability enables a parallel application’s execution system to split or merge processes modifying the parallel application’s granularity. While process migration is widely used to adapt applications to dynamic execution environments, it is limited by the granularity of the application’s processes. Malleability empowers process migration by allowing the application’s processes to expand or shrink following the availability of resources. We have implemented malleability as an extension to the PCM (Process Checkpointing and Migration) library, a user-level library for iterative MPI applications. PCM is integrated with the Internet Operating System (IOS), a framework for middleware-driven dynamic application reconfiguration. Our approach requires minimal code modifications and enables transparent middleware-triggered reconfiguration. We present experimental results that demonstrate the usefulness of malleability.

Department of Computer Science, Rensselaer Polytechnic Institute, Troy, NY

cs-06-17