Trends in HPC Storage
Speaker: Peter Braam
Future exa-scale projects ranging from radio-astronomy and genomics to industrial projects will see unprecedented requirements for storage capacity and throughput. Algorithms to analyse such data are widening from simulation and modelling to include big data analytics. These methods place very different requirements on the storage and compute infrastructures from those seen in traditional HPC and addressed with parallel file systems. While the raw performance seen in HPC environments is not usually found in cloud environments, object storage holds promises and further application focused software and novel hardware will make data intensive problems accessible which seemed out of reach only a few years ago.
Speaker biography:
Dr. Peter Braam is a multidisciplinary innovator covering computing, data and science. As a recognized global leader in computing, without fear to be wrong or incomplete, he thinks systematically to advance technology frontiers and articulates insights in keynotes, white papers and private discussions. Research organizations, governments and enterprises engage with him to create concrete, focused visions and involve him in co-design of solutions. Peter is currently working with Cambridge University on the SKA telescope Science Data Processor and is most known as the creator of the Lustre file system. Prior to founding and running five start-ups (the first four of which were acquired) related to parallel computing and programming languages, Peter was a senior academic at Oxford and Carnegie Mellon
Target audience: Developers and users of supercomputing applications, managers and administrators of HPC data centers
See http://training.cam.ac.uk/event/1440498 for details and to register.