Multicluster architectures overcome the scaling problem of centralized resources by distributing the datapath, register file, and memory subsystem across multiple clusters connec...
In this paper, we describe two new ideas by which HPF compiler can deal with irregular computations e ectively. The rst mechanism invokes a user speci ed mapping procedure via a s...
We describe a radically new cache architecture and demonstrate that it offers a huge reduction in cache cost, size and power consumption whilst maintaining performance on a wide ra...