Alexandre DENISINRIA researcher in the Tadaam research team, formerly in the Runtime research team.
I am working in the field of network communications and middleware for clusters and grids. In particular:
- MPI and communication progression
- multithreaded communications, MPI_THREAD_MULTIPLE
- Intel Xeon Phi
» My publications list on the Runtime pages.
» My publications list in HAL.
NewMadeleine is the latest incarnation of the Madeleine high performance communication library for clusters. It applies optimization strategy on data flows through dynamic packet scheduling, and is usable on various networks such as Infiniband, Myrinet, or Quadrics. I am the official maintainer of NewMadeleine. It comes with a complete MPI interface called Mad-MPI.
» More details are available on the NewMadeleine homepage.
PIOMan is a generic I/O manager, designed to deal with interactions between communication and multi-threading. It guarantees a good level of reactivity, is able to automatically choose between active polling and blocking calls depending on the context, and may offload I/O requests to idle cores when availble to handle multiple requests in parallel and overlap communication and computation. It works with standard pthread or with the Marcel thread scheduler.
» More details are available on the PIOMan homepage.
PadicoTM is a component-based high performance communication framework for grid computing. It enables a wide range of middleware systems (MPI, CORBA, Java RMI, ICE, SOAP,...) to use the variety of networks found in grids.
» More details are available on the PadicoTM homepage.
I am giving:
- Master course on OS and middleware design for parallel and distributed systems at the University Bordeaux.
- Master course on distributed software at ENSEIRB-MATMECA engineering school.
» See my teaching page (french).
Participation to projects
I am involved in the COOP project from the ANR.
I am involved in the France-Japan joint project (JST-CNRS) NEGST project ("NExt Grid Systems and Techniques").
I am involved in the Egide-JSPS Sakura project A multicore-enabled framework for communication and I/O in next generation clusters with the University of Tokyo.
Past activites» Project LEGO
I was involved in the ANR LEGO project ("League for Efficient Grid Operation").
I was involved in the ARC/ACI ALTA project ("Asynchronous Loss Tolerant Algorithms").
I was involved in the INRIA booth at the Super Computing conference.