Contents Preface I Requirements and General Issues Cluster Computing at a Glance Cluster Setup and its Administration Constructing Scalable Services Dependable Clustered Comuting Deploying a High Throughput Computing Cluster Performance Models and Simulation Metacomputing: Harnessing Informal supercomputers Specifying Resources and Services in Metacomputing Systems II Networking, Protocols, and I/O High Speed Networks Lightweight Messaging Systems Active Messages Xpress Transport Protocol Congestion Management in ATM Clusters Load Balancing Over Networks Multiple Path Communication Network RAM Distributed Shared Memory Parallel I/O for Clusters: Methodologies and Systems Software
RAID and Parallel Filesystems III Process Scheduling, Load Sharing, and Balancing Job and Resources Management Systems Scheduling Parallel Jobs on Clusters Load Sharing and Fault Tolerance Manager Parallel Program Scheduling Techniques Customized Dynamic Load Balancing Mapping and Scheduling on Heterogeneous Systems IV Representative Cluster Systems Beowulf RWC PC Cluster II and SCore Cluster System Software COMPaS: A Pentium Pro PC-Based SMP Cluster The NanOS Cluster Operating System BSP-Based Adaptive Parallel Processing MARS: An Adaptive Parallel Programming Environment The Gardens Approach to Adaptive Parallel Computing The ParPar System: A Software MPP Pitt Parallel Computer The RS/6000 SP System: A Scalable Parallel Cluster A Scalable and Highly Available Clustered Web Server Index
