Laurence T. Yang, Minyi Guo
The State of High-Performance Computing Art
Prominent researchers from around the world have focused on presenting the latest techniques and innovations in high-performance computing (HPC), including:
Graph-oriented programming (GOP), OpenMP, stages and transitions (SAT), large-scale synchronous parallel model (BSP), Message Passing Interface (MPI), and Cilk.
Architectural and System Support with techniques for code annotation, application-aware migration with MigThread, check-pointing packages, atomicity-preserving data conversion methods, and lessons learned from applying configurable computing to HPC.
Scheduling and Resource Management Issues, dealing with heterogeneous systems, bus saturation effects on SMP, genetic algorithms for distributed computing, and new task scheduling algorithms.
Clusters and Grid Computing: Design requirements, grid middleware, distributed virtual machines, data grid services, performance enhancement techniques, security issues, and open problems.
Peer-to-Peer Computing: Mechanisms proposed by the High-Performance Fortran (HPF) flood search and routing protocols to improve routing efficiency.
Wireless and Mobile Computing including deploying the Gateway Location Register (GLR) concept in 3G mobile networks, maximizing network lifespan, and comparing QoS-aware scatternet scheduling algorithms.
High-Performance Applications includes partitioning, running Bag-of-Tasks applications on the grid, using low-cost clusters to meet high-demand application requirements, and advanced converged architectures and protocols.
High-Performance Computing: Paradigms and Infrastructure is an invaluable synthesis for engineers, IT experts, computer science and applied mathematics researchers, and students interested in the field.
For more details, visit: https://www.facebook.com/LibUIT.Fanpage/posts/pfbid022y4C5YYNxWUn...
Ha Bang - Media Collaborator, University of Information Technology
Nhat Hien - Translation Collaborator, University of Information Technology
