On cc-NUMA multi-processors, the non-uniformity of main memory latencies motivates the need for co-location of threads and data. We call this special form of data locality, geogra...
DRAM systems achieve high performance when all DRAM banks are busy servicing useful memory requests. The degree to which DRAM banks are busy is called DRAM Bank-Level Parallelism ...
Abstract--Thermal issues are fast becoming major design constraints in high-performance systems. Temperature variations adversely affect system reliability and prompt worst-case de...
Amit Kumar 0002, Li Shang, Li-Shiuan Peh, Niraj K....
Abstract-- We consider the problem of how to improve memory latency tolerance in massively multithreaded GPGPUs when the thread-level parallelism of an application is not sufficien...
Jaekyu Lee, Nagesh B. Lakshminarayana, Hyesoon Kim...
Multiple memory module architecture offers higher performance by providing potentially doubled memory bandwidth. Two key problems in gaining high performance in this kind of archi...