Brad Calder and Dirk Grunwald. Branch prediction architectures for 64-bit address space. Technical Report CU--CS--690--93, Department of Computer Science, University of Colorado at Boulder, 1993.  Home/Search   Document Details and Download   Summary   Related Articles

.... main reasons for this result: within a basic block, successive instructions executed by a processor are stored in successive memory locations; and even though successive basic blocks may not be stored in the memory in sequence, they can be prefetched effectively due to high branch prediction rates [2] [18] 9] On the other hand, achieving very high cache hit rates for data is much tougher to achieve. The reason for such behavior is that the data items consumed by successive instructions need not be stored in successive memory locations. In this paper, we present a novel approach for coupling ....

Brad Calder and Dirk Grunwald. Branch prediction architectures for 64-bit address space. Technical Report CU--CS--690--93, Department of Computer Science, University of Colorado at Boulder, 1993.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC