Title :
MAD kernels: an experimental testbed to study multiprocessor memory system behavior
Author :
Nanda, Arun K. ; Ni, Lionel M.
Author_Institution :
Transarc Corp., Pittsburgh, PA, USA
fDate :
2/1/1996 12:00:00 AM
Abstract :
On large-scale multiprocessors, access to common memory is one of the key performance limiting factors. The shared-memory performance depends not only on the characteristics of the memory hierarchy itself, but also upon the characteristics of the memory address streams and the interaction between the two. We present a technique for multiprocessor workload construction and a family of artificial kernels, called MAD-kernels, to systematically investigate the behavior of the memory hierarchy. The measured performance is independent of any particular application or algorithm. The proposed methodology is demonstrated on two commercial shared-memory systems
Keywords :
performance evaluation; shared memory systems; MAD kernels; interconnection networks; large-scale multiprocessors; memory access patterns; memory hierarchy; multiprocessor memory system; multiprocessor workload construction; performance; performance evaluation; resource contention; shared-memory multiprocessors; shared-memory performance; synchronization overhead; Bandwidth; Degradation; Hardware; Kernel; Large-scale systems; Memory architecture; Multiprocessor interconnection networks; Particle measurements; Performance analysis; System testing;
Journal_Title :
Parallel and Distributed Systems, IEEE Transactions on