Total views : 522

A Comparative Study on CPU Load Predictions in a Computational Grid using Artificial Neural Network Algorithms

Affiliations

  • School of Computing Science and Engineering, VIT University, Vellore – 632014, Tamil Nadu, India
  • School of Electrical Engineering, VIT University, Vellore – 632014, Tamil Nadu, India

Abstract


Background/Objectives: To evaluate the prediction accuracy of Neural Network algorithms for host CPU load prediction and evaluate their performance compared to actual values. Methods/Statistical Analysis: The speed of execution of job at the scheduled host is directly proportional to its CPU load. Therefore, target node load prediction plays an important role in job scheduling decisions. It is learnt that Neural Networks are capable of predicting the future values based on the training given on the past data. We designed a multilayer neural network and trained with learning algorithms for the input patterns collected from the load traces and predicted the future load statistics. The Mean and Standard Deviation of the predicted values are computed and analyzed against the Mean and Standard Deviation of actual values for all the ANN algorithms. Findings: We analyzed the prediction accuracy of Back Propagation, Quick Propagation, Back Propagation with Momentum and Resilient Propagation algorithm for the load traces collected from variety of computers connected in a network. Existing reports shows that Back Propagation algorithm exhibits better prediction accuracy compared to statistical approaches like linear regression and polynomial regression. In this paper, we have shown that Resilient Propagation algorithm has better prediction accuracy compared to other ANN algorithms. Application/Improvements: Job scheduling and resource selection algorithms can employ neural network algorithms to predict the load for the sharable resources connected in the network for more accurate and faster scheduling/resource selection decision.

Keywords

CPU Load Prediction, Load Traces, Neural Network Algorithms, Training, Resource Selection

Full Text:

 |  (PDF views: 408)

References


  • Foster I, Kesselman C. The Grid: Blueprint for a New Computing Infrastructure, Morgan Kaufmann Publishers, San Fransisco, CA, 1998.
  • Yang L, Schopf JM, Foster I. Conservative scheduling: Using predicted variance to improve decisions in dynamic environment.Supercomputing’03. 2003; 31.
  • Jang S, Wu X, Taylor V. Using performance prediction to allocate grid resources. Technical report, GriPhyN. 2004-25; 1–11.
  • Load Trace Archive. Available from: http://www.cs.cmu.edu/~pdinda/LoadTraces/
  • Dinda PA. The statistical properties of host load. Technical report, CMU. 1998; 1–23.
  • Akioka S, Muraoka Y. Extended forecast of CPU and network load on computational grid. IEEE Int’l Symp on Cluster Computing and the Grid. 2004 Apr 19-22; 765–72.
  • Liang J, Nahrstedt K, Zhou Y. Adaptive Multi- Resource Prediction in Distributed Resource Sharing Environment. IEEE Int’l Symp on Cluster Computing and the Grid. 2004 Apr 19-22; 293–300.
  • Dinda PA, O'Hallaron DR. Host load prediction using linear models. Cluster Computing. 2000 Dec; 3(4):265–80.
  • Box G, Jenkins G, Reinsel G. Time Series Analysis Forecasting and Control, Prentice Hall, 1994.
  • Yang L, Foster I, Schopf JM. Homeostatic and tendency-based CPU load predictions. Int’l Parallel and Distributed Processing Symp. (IPDPS'03). 2003 Apr 22-26; 42–50.
  • Dinda PA. A prediction-based real-time scheduling advisor. Proceedings of 16th Int’l Parallel and Distributed Processing Symp (IPDPS 2002). 2002. p. 35–42.
  • Lu D, Sheng H, Dinda P. Size-based scheduling policies with inaccurate scheduling information. 12th IEEE Int’l Symp on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems (MASCOTS'04). 2004 Oct 4-8; 31–8.
  • Swany M, Wolski R. Multivariate resource performance forecasting in the network weather service. Supercomputing’02. 2002; 1–10.
  • Wolski R, Spring N, Hayes J. Predicting the CPU Availability of Time-shared Unix Systems on the Computational Grid. Proceedings of 8th IEEE Symp on High Performance Distributed Computing. 1999; 105–12.
  • Wolski R. Dynamically forecasting network performance using the network weather service. Journal of Cluster Computing.1998 May; 1(1):119–32.
  • Wolski R. Experiences with predicting resource performance on-line in computational grid settings. ACM SIGMETRICS Performance Evaluation Review. 2003 Mar;30(4):41–9.
  • Zebardast B, Maleki I, Maroufi A. A noval multilayer perceptron artificial neural network based recognition for kurdish manuscript. Indian Journal of Science and Technology.2014 Mar; 7(3).
  • Gharehchopogh FS, Khaze SR, Maleki I. A new approach in bloggers classification with hybrid of K-nearest neighbor and artificial neural network algorithms. Indian Journal of Science and Technology. 2015 Feb; 8(3).

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.