Skip to content
Free planNo credit card requiredStart writing
AI for Research home
AI for Research home
Dashboard
Sign inGet started

Search 250 million research papers

Read the abstract, see who wrote it and who cites it, and get the citation in APA, MLA, Chicago or Harvard.

About 10,000 papers for “parallel performance analysis”

Parallel simulation for performance analysis of high performance computer networks

Kris Steenhaut, Wouter Brissinck, K. Degieter et al. · 2002

The study of the characteristics of high performance computer networks is becoming more important as the variety of services offered increases. The basic methods to address this problem are analytical modelling, simulation and measurements. For new or improved systems, only the two first approaches are feasible. It is shown that both suffer from fundamental limitations that hamper their effectiveness in the evaluation of modern high speed networks and transfer modes. A scalable, parallel computer based, solution to the limitations of simulation is presented. It is based on a combination of dedicated hardware and the utilisation of general purpose (MIMD) parallel computers.

Cited by 0

Parallel asynchronous team algorithms: convergence and performance analysis

Benjamı́n Barán, Eugenius Kaszkurewicz, Amit Bhaya · IEEE Transactions on Parallel and Distributed Systems · 1996

This paper formalizes a general technique to combine different methods in the solution of large systems of nonlinear equations using parallel asynchronous implementations on distributed-memory multiprocessor systems. Such combinations of methods, referred to as team algorithms, are evaluated as a way of obtaining desirable properties of different methods and a sufficient condition for their convergence is derived. The load flow problem of electrical power networks is presented as an example problem that, under certain conditions, has the characteristics to make a team algorithm an appealing choice for its solution. Experimental results of an implementation on an Intel iPSC/860 Hypercube are reported, showing that considerable speedup and robustness can be obtained using team algorithms.

Cited by 40

Traveler: Navigating Task Parallel Traces for Performance Analysis

Sayef Azad Sakin, Alex Bigelow, Rezaei Tohid et al. · IEEE Transactions on Visualization and Computer Graphics · 2022

Understanding the behavior of software in execution is a key step in identifying and fixing performance issues. This is especially important in high performance computing contexts where even minor performance tweaks can translate into large savings in terms of computational resource use. To aid performance analysis, developers may collect an execution trace-a chronological log of program activity during execution. As traces represent the full history, developers can discover a wide array of possibly previously unknown performance issues, making them an important artifact for exploratory performance analysis. However, interactive trace visualization is difficult due to issues of data size and complexity of meaning. Traces represent nanosecond-level events across many parallel processes, meaning the collected data is often large and difficult to explore. The rise of asynchronous task parallel programming paradigms complicates the relation between events and their probable cause. To address these challenges, we conduct a continuing design study in collaboration with high performance computing researchers. We develop diverse and hierarchical ways to navigate and represent execution trace data in support of their trace analysis tasks. Through an iterative design process, we developed Traveler, an integrated visualization platform for task parallel traces. Traveler provides multiple linked interfaces to help navigate trace data from multiple contexts. We evaluate the utility of Traveler through feedback from users and a case study, finding that integrating multiple modes of navigation in our design supported performance analysis tasks and led to the discovery of previously unknown behavior in a distributed array library.

Cited by 9Open access

Integrating Power and Cooling into Parallel Performance Analysis

Rashawn L. Knapp, Karen L. Karavanic, Andrés Márquez · 2010

We present a new, integrated approach to parallel performance analysis that integrates traditional application-oriented performance data with measurements of the physical runtime environment. We have developed the needed infrastructure for combined evaluation of system, application, and machine room performance in the high end environment. We illustrate the utility of our approach, with data from our study of the power and cooling impact of the choice of physical location for an application within the machine room. We demonstrate the integration of measured performance data from the application, system, and physical room environment, and discuss the challenges encountered.

Cited by 1

Performance Analysis of Parallel Systems: A Hybrid Approach

Rector Magni, Prof. Dr. J. Wessels, Prof. Dr. N. M et al. · 2008

proefschrift ter verkrijging van de graad van doctor aan de Technische Universiteit Delft, op gezag van de Rector Magni cus Prof.ir. K.F. Wakker, in het openbaar te verdedigen ten overstaan van een commissie, door het College van Dekanen aangewezen, op maandag 2 oktober 1995 te 16.00 uur door Hendrik JONKERS ingenieur informatica geboren te Amsterdam Dit proefschrift is goedgekeurd door de promotor:

Cited by 0

Improving parallel program performance using critical path analysis

Andrew W. Kwan, Lubomir Bic, Daniel D. Gajski · eScholarship (California Digital Library) · 1989

A programming tool that performs analysis of critical paths for parallel programs has been developed. This tool determines the critical path for the program as scheduled onto a parallel computer with P processing elements, the critical path for the program expressed as a data flow graph (when maximal parallelism can be expressed), and the minimum number of processing elements (P_opt) needed to obtain maximum program speedup. Experiments were performed using several versions of a Gaussian elimination program to examine how speedup varied with changes in granularity and critical path length. These experiments showed that when the available numer of processing elements P < P_opt, increasing granularity improved program speedup more than reducing (the data flow graph's) critical path length, whereas when P ≥ P_opt, increasing granularity degraded program speedup while reducing critical path length improved program speedup.

Cited by 3Open access

Performance Analysis of Mapping Strategies for Parallel Simulation

2008

This paper presents an analytical model to evalu-ate the performance of conservative pamllel simulation using a finite-buffered multistage interconnection net-work as an example. The model is formalized based on two important time components in parallel and dis-tributed processing: computation time and communi-cation time. Our mathematical analysis identifies the major constituents of simulation overheads that affect the performance of pamllel simulation. We also show that a perfectly balanced workload distribution may not necessarily translates into better performance. On the contmry, we have shown that a balanced mapping of workload may increase communication overheads result-ing in a longer simulation elapsed time.

Cited by 0

Holistic Hardware Counter Performance Analysis of Parallel Programs

Brian J. N. Wylie A, Bernd Mohr A, Felix Wolf A · 2008

The KOJAK toolkit has been augmented with refined hardware performance counter support, including more convenient measurement specification, additional metric derivations and hierarchical structuring, and an extended algebra for integrating multiple experiments. Comprehensive automated analysis of a hybrid OpenMP/MPI parallel program is demonstrated with performance experiments, containing communication and synchronisation metrics combined with a rich set of counter metrics, which provide a holistic analysis context and facilitate multi-platform comparison. 1.

Cited by 0

Performance Analysis of Image Segmentation Using Parallel Processing

International Journal of Innovative Research in Computer Science and Technology (IJIRCST) · Bulletin of Miscellaneous Information (Royal Gardens Kew) · 2015

This paper presents a study on Fuzzy C Means (FCM) based segmentation technique via different modes over a number of images. This research work includes development of simulation models in both sequential and parallel modes. The proposed FCM based segmentation method is tested extensively by subjective and objective evaluation via both modes. Various image and performance parameters are evaluated such as serial time, parallel time, overheads, efficiency, MMSE, PSNR and Percentage of Tumour. The results show that leveraging of multi-core platforms can speed up the processing of images considerably through the use of parallel computing tools in MATLAB. The parameters like PSNR, MMSE, and image quality have not been compromised while increasing the time efficiency through parallel mode. This information can be utilized in the fields where real time processing of image data is required.

Cited by 0

SPMD Performance Analysis with Parallel Computing of Matlab

Junjun Cao, Shanshan Fan, Xuan Yang · 2012

Matlab Parallel Toolbox is an effective tool for developing parallel program under a multi-processor environment, in which SPMD is the key statement for single instruction multiple data parallel implementation. However, when using spmd to achieve parallel program, beginners may find that it can hardly improve the speedup of program, and is even worse than the serial one. In this paper, we analyze the matrix operation, for-loop control and the influence the computing scale has on the performance of SPMD, and explain why SPMD speedup is hard to be improved through plenty of experiments. Finally we propose effective methods using spmd to improve parallel program speedup and give sample code of joint histogram on image application.

Cited by 5

Developing a scalable, extensible parallel performance analysis toolkit

Adam Bordelon · Rice Digital Scholarship Archive (Rice University) · 2007

Modern parallel systems and applications are constantly increasing in scale and complexity, and consequently good parallel performance is impossible to achieve without the help of performance tools. However, monitoring application performance on these large-scale systems generates massive amounts of performance data. Current performance tools are insufficient for practical analysis of such large-scale data, typically either showing only basic summary information, or bombarding the user with all of the performance details with little help for pinpointing useful patterns. This thesis presents HPCVision, an extensible tool framework with a novel approach for scalable parallel performance analysis and visualization. This framework provides two performance toolkits for examining similarities and differences in parallel performance among an ensemble of processes, identifying equivalence classes of behavior, and pinpointing performance anomalies. HPCVision presents the performance data and analysis results in an intuitive, scalable manner to provide insight into application performance, automating the tuning cycle and increasing the productivity of the human analyst.

Cited by 2Open access

Singularity analysis and performance indexes for parallel manipulators

Giovanni Boschetti, Alessandro Gasparetto, V. Zanotto · Research Padua Archive (University of Padua) · 2008

The performance indexes are commonly based on the Jacobian Matrix, and they have been commonly used in determining the dexterous regions of a manipulator workspace. Many performance indexes have been studied so far for the kinematic behavior analysis of serial manipulators, and most of them can be applied to the parallel ones too. The Manipulability Index, the Minimum Singular Value and the Reverse of the Condition Number give information about the presence of points of singularity in the manipulator workspace. In the first part of the paper, the attention is focused in the problem of the singularity analysis of the parallel manipulators by means of the most commonly used performance indexes. A manipulator with three degrees of freedom will be used as a testbed for this analysis. In the second part, the paper deals with novel performance indexes that have the property of giving information even in the points of singularity of the manipulator workspace.

Cited by 0

Performance analysis of massively-parallel computational fluid dynamics

James Hawkes, Stephen R. Turnock, Simón Cox et al. · ePrints Soton (University of Southampton) · 2014

As modern supercomputers edge towards exascale, their architectures are becoming more parallel. In order for computational fluid dynamics (CFD) simulations to operate efficiently on newer machines, a complete harmony between hardware, software and numerical algorithms is required. In the work presented here, a typical CFD code is instrumented, and a strong-scalability study performed to identify areas of the execution which require improvement, using the well-known KVLCC2 test case. The effects of changing discretization schemes, mesh structure, turbulence models and linear solvers are all tested. The results show that data exchange among cores and the inner-loop pre-conditioners both have a large impact on performance in a massively-parallel environment, and should be the focus of future developments.

Cited by 8Open access

Performance Analysis of Parallel Apriori on Heterogeneous Nodes

Ketan Shah, Sunita Mahajan · 2009

The paper analyzes the performance of parallel apriori algorithm on heterogeneous nodes with different datasets and over n processors on a commodity cluster of machines. In the apriori algorithm all processes need to synchronize after every pass. If any process is assigned more load than other processes in the system, the slowest process will dictate the speed of the program. It is therefore important to ensure that load is equally balanced among all processes. Memory, speed of the processor and cache play a significant role in the processing capacity of the system. The experiments show that nodes with different configurations affect the performance of the parallel apriori algorithm. In order to maximize the efficiency it is required to balance the data set based on the processing speed of the various nodes present in the cluster.

Cited by 0