High Speed and Large Scale Scientific Computing
Wolfgang Gentzsch, Lucio Grandinetti, Gerhard Robert Joubert
IOS Press, 2009 - Computers - 486 pages
During the last decade parallel technologies have completely transformed main stream computing. The majority of standard PC 's today incorporate multi-processor chips, with up to four processors. This number will reach eight and more processors soon. The flexibility that parallel systems constructed with commodity components offer, make them easy to link through wide area networks, such as the Internet, to realize Grids or Clouds.The immediate benefit of these networks is that they can be accessed by a wide community of users, from many different disciplines, to solve compute intensive or data intensive problems requiring high speed computing resources.'High Speed and Large Scale Scientific Computing' touches upon issues related to the new area of Cloud computing, discusses developments in Grids, Applications and Information Processing, as well as e-Science.
What people are saying - Write a review
algorithm Amazon analysis Aneka Aneka Cloud applications architecture bandwidth blocks business models Cilk cloud computing collaboration communication components computing cluster computing resources configuration core cost CUDA data regeneration data transfer DEISA deployed deployment distributed dynamic e-Science EGEE environment ESnet European example execution experiments Figure FPGAs framework function global Grid Computing grid infrastructure Hadoop hardware High Performance high-performance computing IEEE implementation InfiniBand input integrated interface IOS Press iteration KISTI layer license loop MapReduce matrix messages metadata middleware monitoring multicore multiple OpenNebula operations optimization overhead parallel Parallel Computing physical platform plug-in problems processors programming model provides query remote replica requirements running runtime S-Net scalability scheduling scientific server simulation solution specific storage supercomputing tasks threads tile UNICORE virtual machines workflow