High-end supercomputers are increasingly built out of commodity components, and lack tight integration between the processor and network. This often results in inefficiencies in ...
Christian Bell, Dan Bonachea, Yannick Cote, Jason ...
High-performance caches statically pull up the bitlines in all cache subarrays to optimize cache access latency. Unfortunately, such an architecture results in a significant wast...
We present new architectural concepts for uniprocessor designs that conform to the data-driven computation paradigm. Usage of our D2 -CPU (Data-Driven processor) follows the natura...
A central problem in executing performance critical parallel and distributed applications on shared networks is the selection of computation nodes and communication paths for exec...
Cryptographic algorithms are more efficiently implemented in custom hardware than in software running on general-purpose processors. However, systems which use hardware implementat...