Communications of the ACM
Image template matching on MIMD hypercube multicomputers
Journal of Parallel and Distributed Computing
Scalability of parallel machines
Communications of the ACM
Computing biconnected on a hypercube
The Journal of Supercomputing
Introduction to parallel algorithms and architectures: array, trees, hypercubes
Introduction to parallel algorithms and architectures: array, trees, hypercubes
Scalable problems and memory-bounded speedup
Journal of Parallel and Distributed Computing
Analyzing scalability of parallel algorithms and architectures
Journal of Parallel and Distributed Computing - Special issue on scalability of parallel algorithms and architectures
Anomalies in parallel branch-and-bound algorithms
Communications of the ACM
Scalability of Parallel Algorithm-Machine Combinations
IEEE Transactions on Parallel and Distributed Systems
Shared Virtual Memory and Generalized Speedup
Proceedings of the 8th International Symposium on Parallel Processing
Integrated Range Comparison for Data-Parallel Compilation Systems
IEEE Transactions on Parallel and Distributed Systems
On the Influence of Start-Up Costs in Scheduling Divisible Loads on Bus Networks
IEEE Transactions on Parallel and Distributed Systems
Use of a CORBA/RMI gateway: characterization of communication overhead
WOSP '02 Proceedings of the 3rd international workshop on Software and performance
Performance-steered design of software architectures for embedded multicore systems
Software—Practice & Experience
Scalability versus execution time in scalable systems
Journal of Parallel and Distributed Computing
Performance Prediction: A Case Study Using a Scalable Shared-Virtual-Memory Machine
IEEE Parallel & Distributed Technology: Systems & Technology
IEEE Computational Science & Engineering
Evaluation of Parallel Programs by Measurement of Its Granularity
PPAM '01 Proceedings of the th International Conference on Parallel Processing and Applied Mathematics-Revised Papers
Fine-grain design space exploration for a cartographic SoC multiprocessor
ACM SIGARCH Computer Architecture News
Speedup and scalability analysis of Master--Slave applications on large heterogeneous clusters
Journal of Parallel and Distributed Computing
A pilot study to compare programming effort for two parallel programming models
Journal of Systems and Software
High level performance metrics for FPGA-based multiprocessor systems
Performance Evaluation
Mathematical and Computer Modelling: An International Journal
Hi-index | 0.00 |
Parallel computing's much-heralded triumph has failed to arrive with all its anticipated thunder. This disappointing acceptance of parallel computing springs from several reasons, the first three technological and the fourth largely economic: Lack of a unifying model. Parallel computing has no simple, acceptably accurate model whose algorithms run as well on the model as on a real parallel computer. Lack of program portability. To change parallel computers usually requires that users rewrite or at least retune all programs according to a number of features. Lack of suitable performance metrics. Performance metrics for parallel algorithms consequently are tied to the target parallel architecture, and there are as many of these algorithm-architecture combinations as there are different parallel architectures. Use of slow processors. Parallel computers frequently use serial processors that are significantly slower than the fastest PCs and workstations, making it difficult to show spectacular gains over the latest serial competitors. The authors review the various proposed metrics to discover why so many performance metrics for parallel systems currently exist. They then show that while the focus of much recent research has shifted to optimizing performance metrics, runtime should remain the primary measure. Elevating almost any other metric to the primary position runs the risk of favoring a parallel algorithm that always runs slower over one that always runs faster.