Annual Review of Scalable ComputingA collection of seven long articles, this book comprehensively discusses significant projects in scalable computing in various research organizations around the world. It represents the quantitative and qualitative growth of work in the area. Contents: Experiences with Shared Virtual Memory on System Area Network Clusters: System Simulation, Implementation, and Emulation; Average-Case Scalability Analysis of Parallel Computations; Parallel IO Prefetching and Caching; A C++/Tuple-Lock Implementation for Distributed Objects; Static Data Allocation and Load Balancing Techniques for Heterogeneous Systems; Building a Global Object Space for Supporting Single System Image on a Cluster; A Computation-Centric Multilocation Consistency Model for Shared Memory. Readership: Graduate students, academics and researchers in supercomputing and computer engineering. |
Contents
System Simulation Implementation and Emulation | 1 |
2 AverageCase Scalability Analysis of Parallel Computations | 51 |
3 Parallel IO Prefetching and Caching | 96 |
4 A C++TupleLock Implementation for Distributed Objects | 152 |
5 Static Data Allocation and Load Balancing Techniques for Heterogeneous Systems | 182 |
6 Building a Global Object Space for Supporting Single System Image on a Cluster | 225 |
7 A ComputationCentric Multilocation Consistency Model for Shared Memory | 261 |
Other editions - View all
Common terms and phrases
acquire and release acyclic allocation applications assigned average-case efficiency Backer balanced phase bandwidth buffer cache cluster communication competitive ratio Computer Architecture consistency model diffs Distributed Shared Memory edge evicted execution fetch Figure function GeNIMA Hence heterogeneous hot disk I/O schedule IEEE implementation interrupt cost isospeed scalability J. P. Singh Java JESSICA L-block lookahead L-OPT Lemma loaded phase lookahead memory coherence messages migration modify mprotect multiprocessors Myrinet network interface number of blocks number of I/Os online algorithm operating system OPT(k optimal overhead page faults parallel algorithm parallel computation parallel I/O parameters partitioning PC-OPT performance Plog prefetching priority problem protocol processing queue rectangles reference string referenced release consistency remote requests scheduling algorithm Section sequential consistency shared memory shared object shared object handler shared virtual memory speedup Symposium synchronization tasks Theorem thread TOPT total number tuple tuple-lock updates write