Hermes

Accelerating the Performance of Multi-Site Scientific applications through Coordinated Data management

Hermes is an associated team between the Myriads team from INRIA Rennes-Bretagne Atlantique and Lawrence Berkeley National Lab (LBNL).

Principal investigator (Inria): Shadi Ibrahim, Myriads team.
Principal investigators (partner): Suren Byna, Lawrence Berkeley National Lab, USA

Partners

Myriads Team, Inria Rennes - Bretagne Atlantique, France.

Shadi Ibrahim, Research Scientist within the Myriads team. His research focuses on scal- able big data management in Clouds and HPC systems. He is currently leading the ANR KerStream project which develops new middleware for stream data processing ap- plications and the Apollo connect talent project which targets scalable data management in geo-distributed environments; and is a member the DISCOVERY Inria Project Lab and JLESC: Joint Laboratory on Extreme-Scale Computing. He has served as a PC member for international conferences in cloud, HPC and parallel and distributed systems (ACM/IEEE SC, IEEE ICPP, IEEE Cluster, ACM/IEEE CCGrid), and as Program co- chair for the 2018 IEEE International Congress on Big Data, the 2017 ICA3PP conference and two workshops on the convergence of Big Data and HPC (CEBDA@IPDPS2018 and EBDMA@CCGrid 2017), and as guest editors for two journal special issues in IEEE Transactions on Big data and FGCS.

Thomas Lambert, is a postdoctoral researcher within the Myriads Team. Thomas is funded by the ANR KerStream project. His research focuses on resource management and scheduling for Big Data applications in heterogeneous environments.

David Guyon, was a postdoctoral researcher within the STACK Team. David is funded by the Connect Talent project. His research focuses on data management in distributed Clouds, Stream data processing and Energy-Efficiency in heterogeneous environments.

Jad Darrous, was a Ph.D. student co-advised by Christian Perez (Inria, LIP) and Shadi Ibrahim, since October 2016. Jad is funded by Inria within the framework of the IPL Discovery project. His research interests include data management in Geo-distributed Clouds.

Scientific Data Management Group (SDM) at Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA, USA

Suren Byna, is a Staff Scientist in the SDM Group at LBNL. His broad research interests are in large-scale scientific data management. Specifically, his research focus is in opti- mizing parallel I/O and in developing systems for efficient movement of scientific data. He currently leads several projects funded by the Office of Science in the US Department of Energy (DOE). He is the PI of the Exascale Computing Projects (ECP) ExaHDF5 project, which develops production-ready features in HDF5. He leads projects to de- sign next-generation storage systems aimed at scalable object-centric data management systems and to develop data management strategies for supporting massive amounts of ex- perimental and observation data (EOD). He serves on the program committees of a large number of conferences in the HPC field, such as Supercomputing, IPDPS, IEEE Clus- ter, ACM/IEEE CCGrid, and Journals of Parallel and Distributed Computing (JPDC), Supercomputing, and Cluster Computing.

Alex Sim, is a Senior Computing Engineer in the SDM group at LBNL. He has interests in wide-area networking, resource management, data flow modeling, data mining, and machine learning. He currently works on multi-dimensional, high frequency streaming data analysis algorithms. He has worked on Open Science Grid, DOE SciDAC projects, Earth System Grid Federation (ESGF), Storage Resource Management (SRM), Particle Physics Data Grid, and two Next Generation Internet projects.

Bin Dong, is a Young Research Scientist in the SDM Group, at LBNL. His research work focuses on developing systems for using burst buffers in HPC systems, for querying large- scale scientific data, and for analyzing data efficiently. His research is funded by the ECP ExaHDF5 and the Scientific Data Services (SDS) project.