SPECIAL SESSION CHAIR

Cloud computing, distributed HPC systems, and future networks

Dr. Drissa Houatra
Orange Labs, France

Abstract

Cloud computing infrastructure services rely largely on HPC technologies. In particular, recent developments in multicore and graphical processors, high-capacity storage systems, supercomputers and cluster servers have led to the possibility to develop and interconnect large datacenters which we view as distributed high-performance infrastructure services for cloud computing. The consolidation and evolution of these infrastructure services will lead to revisiting some of the hardware and system software solutions in use in parallel and distributed computing research and in the ICT industry (e.g. LAN interconnect solutions, access to very large files and databases, virtualized parallel processing environments). Most importantly, by exploiting synergies between several ICT domains (HPC and high-speed networks, grid computing, autonomics, middleware and SOA, as well as virtualization), cloud computing enables novel forms of distributed services and global networking applications. Classical HPC applications in e-Science and the industry will increasingly rely on cloud service offers running on distributed datacenters as a solution to their computing resource shortage, and perhaps most importantly, problems related to the operations of their computing assets, which are becoming a nightmare. Also, some of the applications enabled by cloud computing will probably need to leverage HPC techniques developed by the scientific and industrial simulation community in order to realize their full potential. Future corporate data networks, IP networks or Web-based computer networks in particular will be shaped by cloud infrastructure services. These networks are positioned as a key enabler for the development of geographically distributed datacenters and the deployment of cloud applications at the global scale. This special session is a framework to develop new design concepts or engineering solutions arising from synergies between computer cloud concepts, high-performance network architectures and distributed computing services.

Biography of the Special Session Chair

Dr. Drissa Houatra is a distributed systems and networking researcher working at Orange Labs, Paris (Issy), France. He received a Ph.D. in computer science (parallel and distributed systems) from the National Polytechnic Institute of Toulouse (INPT), France, in 1994. After his doctorate, he began work by gaining some teaching experience as a temporary associate professor at the Ecole Nationale Supérieure d'Electrotechnique, d'Electronique, d'Informatique, d'Hydraulique et des Télécommunications (ENSEEIHT), Toulouse, an engineering school. Since 1996, he has been conducting research at CNET France Telecom (now Orange Labs) on a variety of topics, including telecommunication information networking architectures, CORBA-based intelligent networks, distributed objects and real-time multimedia, distributed virtual continuums and avatar models, programmable networks and policy-based management systems, autonomics, grid computing and virtualisation, cloud networks and the evolution of telecom datacenters to support cloud services. He is also a networking instructor at the University of Paris 11 (Orsay). He has co-authored more than 60 scientific and technical publications, mostly conference papers and research reports (including work recognized with a "best paper award"), and also journal papers, book chapters, patent descriptions, as well as project proposals.

Submissions

Please email all submissions to calgary@iasted.org by August 25, 2010. Authors MUST include their full contact information in the email.