Highly Parallel Computing
Written by an expert in the field of parallel processing, this revised edition discusses both hardware and software issues involved in designing parallel systems in today's architectures. An explanation of the fundamental concepts of synchronization is followed by a sample of applications from engineering and the sciences, including an informative description of the parallel model in database management.
What people are saying - Write a review
We haven't found any reviews in the usual places.
Technological Constraints and Opportunities
9 other sections not shown
Other editions - View all
algorithm applications architecture array bandwidth basic blocks bits buffer cache called chapter chip circuit communication compiler complexity computational model concurrent connected Cosmic Cube Cray X-MP critical section cycle data structures database dataflow languages described discussed disk dynamic efficient evaluation example fetch-and-add floating-point Fortran functional language global graph hardware hypercube ILLIAC IV implementation input instruction interconnection iterations latency lazy evaluation logic loop memory modules message-passing MFLOPS microcode MIMD multiple multiprocessor needed Neumann node number of processors omega network operating system output parallel computer parallel execution parallel processing parallel program performance pipeline problem processing elements programming languages Prolog queue registers scheduling semaphores serial shared shared-memory shown in Figure SIMD simulation single speedup static storage switch synchronization systolic array task tokens topology transistor tree Ultracomputer UNIX update variable vector processor VLSI