Estimating interlock and improving balance for pipelined architectures
Journal of Parallel and Distributed Computing
Memory contention for shared memory vector multiprocessors
Proceedings of the 1992 ACM/IEEE conference on Supercomputing
Performance and Scalability of Preconditioned Conjugate Gradient Methods on Parallel Computers
IEEE Transactions on Parallel and Distributed Systems
Public international benchmarks for parallel computers: PARKBENCH committee: Report-1
Scientific Programming
Predictive performance and scalability modeling of a large-scale application
Proceedings of the 2001 ACM/IEEE conference on Supercomputing
Proceedings of the 2003 ACM/IEEE conference on Supercomputing
Proceedings of the 2006 ACM/IEEE conference on Supercomputing
A note on scaling the Linpack benchmark
Journal of Parallel and Distributed Computing
Computational forces in the Linpack benchmark
Journal of Parallel and Distributed Computing
Computational forces in the SAGE benchmark
Journal of Parallel and Distributed Computing
Identifying sources of Operating System Jitter through fine-grained kernel instrumentation
CLUSTER '07 Proceedings of the 2007 IEEE International Conference on Cluster Computing
Dimensional analysis applied to a parallel QR algorithm
PPAM'07 Proceedings of the 7th international conference on Parallel processing and applied mathematics
Impact of noise on scaling of collectives: an empirical evaluation
HiPC'06 Proceedings of the 13th international conference on High Performance Computing
The impact of noise on the scaling of collectives: a theoretical approach
HiPC'05 Proceedings of the 12th international conference on High Performance Computing
Computer performance analysis and the Pi Theorem
Computer Science - Research and Development
Hi-index | 0.00 |
Self-similarity is a property of physical systems that describes how to scale parameters such that dissimilar systems appear to be similar. Computer systems are self-similar if certain ratios of computational forces, also known as computational intensities, are equal. Two machines with different computational power, different network bandwidth and different inter-processor latency behave the same way if they have the same ratios of forces. For the parallel conjugate gradient algorithm studied in this paper, two machines are self-similar if and only if the ratio of one force describing latency effects to another force describing bandwidth effects is the same for both machines. For the two machines studied in this paper, this ratio, which we call the mixing coefficient, is invariant as problem size and processor count change. The two machines have the same mixing coefficient and belong to the same equivalence class.