What people are saying - Write a review
We haven't found any reviews in the usual places.
ILP and WideBusses
41 other sections not shown
adaptive algorithm application array bandwidth barrier benchmarks binomial tree block branch branch prediction buffer cache line cache misses collective I/O communication compiler copy cost CP-PACS cycle data cache data transfer data transformations dependences described disk disk array distributed Distributed Shared Memory ENDDO equations evaluation example execution Fortran function global hardware implementation improvement increase instructions Intel Paragon interface iteration space latency load/store loop nest loop transformations machine mapping matrix memory access method migratory miss ratio multiple multiprocessor node number of processors operations optimal overhead page faults parallel computers parallel programming parameters partitioning path performance prediction prefetching problem Proc protocol query requests requires runahead scheduling scheme Section server shared memory shown in Figure shows simulation speedup stall stride subtree superscalar synchronization Table techniques template tion traffic transformations tree update variable vector processors victim cache VLIW