Realistic parallel performance estimation

Hey, Tony; Dunlop, Alistair N.; Hernández, Emilio

doi:10.1016/s0167-8191(96)00093-2

Cited by 12 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is a multivariate analysis that tries to characterize the relations among the data. The difference with current work, such as PMaC [28] or PERFORM [12], is that they work with relational structures that are a priori chosen or have to be configured manually.…”

Section: Related Workmentioning

confidence: 99%

“…The first condition of Eq. (12) implies that X and Z should be related. The second condition states that X is only indirectly related to Z via Y .…”

Section: Deterministic Relationsmentioning

confidence: 99%

“…The adjacency search procedure was modified for finding equivalent variables by comparing the conditional independence tests according to Eq. (12). An extra step was added to the algorithm in which the complexity of equivalent relations is calculated using a regression analysis and the one with the least complexity is chosen.…”

Section: Deterministic Relationsmentioning

confidence: 99%

See 2 more Smart Citations

Causal Analysis for Performance Modeling of Computer Programs

Lemeire

Dirkx

Verbist

2007

Scientific Programming

View full text Add to dashboard Cite

Causal modeling and the accompanying learning algorithms provide useful extensions for in-depth statistical investigation and automation of performance modeling. We enlarged the scope of existing causal structure learning algorithms by using the form-free information-theoretic concept of mutual information and by introducing the complexity criterion for selecting direct relations among equivalent relations. The underlying probability distribution of experimental data is estimated by kernel density estimation. We then reported on the benefits of a dependency analysis and the decompositional capacities of causal models. Useful qualitative models, providing insight into the role of every performance factor, were inferred from experimental data. This paper reports on the results for a LU decomposition algorithm and on the study of the parameter sensitivity of the Kakadu implementation of the JPEG-2000 standard. Next, the analysis was used to search for generic performance characteristics of the applications.

show abstract

Section: Related Workmentioning

confidence: 99%

“…The first condition of Eq. (12) implies that X and Z should be related. The second condition states that X is only indirectly related to Z via Y .…”

Section: Deterministic Relationsmentioning

confidence: 99%

See 1 more Smart Citation

Causal Analysis for Performance Modeling of Computer Programs

Lemeire

Dirkx

Verbist

2007

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…Even at this intermediate level though, all previous modelling techniques, such as those proposed by Adve [1], Mehra et al [20], Parashar and Hariri [24], Jonkers [16], van Gemund [27], Labarta et al [18] and Dunlop and Hey [12] fail to strike an ideal balance between accuracy, flexibility, implementation and evaluation costs (see Grove's thesis [9] for a detailed discussion of the advantages and disadvantages of each of these approaches). Even at this intermediate level though, all previous modelling techniques, such as those proposed by Adve [1], Mehra et al [20], Parashar and Hariri [24], Jonkers [16], van Gemund [27], Labarta et al [18] and Dunlop and Hey [12] fail to strike an ideal balance between accuracy, flexibility, implementation and evaluation costs (see Grove's thesis [9] for a detailed discussion of the advantages and disadvantages of each of these approaches).…”

Section: Previous Work On Performance Modellingmentioning

confidence: 99%

Communication Benchmarking and Performance Modelling of MPI Programs on Cluster Computers

Grove

Coddington

2005

J Supercomput

View full text Add to dashboard Cite

This paper gives an overview of two related tools that we have developed to provide more accurate measurement and modelling of the performance of message-passing communication and application programs on distributed memory parallel computers. MPIBench uses a very precise, globally synchronised clock to measure the performance of MPI communication routines. It can generate probability distributions of communication times, not just the average values produced by other MPI benchmarks. This allows useful insights to be made into the MPI communication performance of parallel computers, and in particular how performance is affected by network contention. The Performance Evaluating Virtual Parallel Machine (PEVPM) provides a simple, fast and accurate technique for modelling and predicting the performance of message-passing parallel programs. It uses a virtual parallel machine to simulate the execution of the parallel program. The effects of network contention can be accurately modelled by sampling from the probability distributions generated by MPIBench. These tools are particularly useful on clusters with commodity Ethernet networks, where relatively high latencies, network congestion and TCP problems can significantly affect communication performance, which is difficult to model accurately using other tools. Experiments with example parallel programs demonstrate that PEVPM gives accurate performance predictions on commodity clusters. We also show that modelling communication performance using average times rather than sampling from probability distributions can give misleading results, particularly for programs running on a large number of processors.

show abstract

“…The PERFORM system developed by Dunlop at Southampton is an execution-driven simulation tool that uses a novel 'Fast Simulation Method' that attempted to improve the accuracy of prediction and overcome some of the problems with a full simulation of the memory hierarchy [10,14]. The model uses the "program slicing" technique to isolate the control variables and array indices of the source code, retaining sufficient information to simulate data movement within the memory hierarchy.…”

Section: Simulationmentioning

confidence: 99%

Performance Engineering, PSEs and the GRID

Hey

Papay

2002

Scientific Programming

Self Cite

View full text Add to dashboard Cite

Performance Engineering is concerned with the reliable prediction and estimation of the performance of scientific and engineering applications on a variety of parallel and distributed hardware. This paper reviews the present state of the art in 'Performance Engineering' for both parallel computing and meta-computing environments and attempts to look forward to the application of these techniques in the wider context of Problem Solving Environments and the Grid. The paper compares various techniques such as benchmarking, performance measurements, analytical modelling and simulation, and highlights the lessons learned in the related projects. The paper concludes with a discussion of the challenges of extending such methodologies to computational Grid environments.

show abstract

Realistic parallel performance estimation

Cited by 12 publications

References 17 publications

Causal Analysis for Performance Modeling of Computer Programs

Causal Analysis for Performance Modeling of Computer Programs

Communication Benchmarking and Performance Modelling of MPI Programs on Cluster Computers

Performance Engineering, PSEs and the GRID

Contact Info

Product

Resources

About