[1]XIE Yi,TANG Chenghua,HUANG Xiangnong.A doubly hidden Markov model for synthesizing bursty workloads[J].CAAI Transactions on Intelligent Systems,2012,7(2):108-114.
Copy

A doubly hidden Markov model for synthesizing bursty workloads

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 7 Number of periods: 2012 2 Page number: 108-114 Column: 学术论文—人工智能基础 Public date: 2012-04-25
Title:
A doubly hidden Markov model for synthesizing bursty workloads
Author(s):
XIE Yi1;  TANG Chenghua2;  HUANG Xiangnong3
1. School of Information Science and Technology, Sun YatSen University，Guangzhou 510006, China;
2. School of Computer Science and Engineering, Guilin University of Electronic Technology, Guilin 541004, China;
3. Network and Information Technology Center, Sun YatSen University， Guangzhou 510275, China
Keywords:
hidden Markov model;  synthesize;  burst workload;  network
CLC:
TP30
DOI:
-
Abstract:
Network traffic models have been widely used to build the test environment for networks and network services. Their accuracy directly impacts the performance evaluation results of various services and their robustness in the actual network environment. With the popularity of ecommerce and new network applications, the burst traffic phenomenon has become one of the main features of the modern internet. Traditional traffic models designed for stationary network traffic have difficulty in effectively describing the temporal structure and statistical properties of burst traffic of modern networks, which causes them not to be able to accurately reflect the actual network environment. In this paper, a new structural doubly hidden Markov model was proposed to characterize the practical burst traffic in a real network environment. Efficient algorithms for inference of model parameters and synthesis of the burst workload were also introduced. Based on the hierarchical structure, the proposed model can reproduce the similar temporal structure, statistical properties, and selfsimilarity of the real burst traffic. The proposed model includes two hidden Markov processes. The parent Markov state process was used to describe the largescale trends or phases of burst traffic. The child Markov process was used to describe the smallscale fluctuations that happen during a given phase of the arrival process. Experiments were implemented to validate the proposed model.
References:
［1］PAXSON V, FLOYD S. Wide area traffic: the failure of Poisson modeling［J］. IEEE/ACM Transactions on Networking (ToN), 1995, 3(3): 244.
［2］LELAND W, TAQQU M, WILLINGER W, et al. On the selfsimilar nature of Ethernet traffic［J］. IEEE/ACM Transactions on Networking, 2002, 2(1): 115. 
［3］BERAN J, SHERMAN R, TAQQU M, et al. Long range dependence in variable bit rate video traffic［J］. IEEE Trans on Communication, 1995, 43: 15661579.
［4］ZVIRONIENE A, NAVICKAS Z, RINDZEVICIUS R. Bursty traffic simulation by ONOFF model［J］. Electronics and Electrical Engineering, 2006, 70(6): 6568. 
［5］RIEDI R, CROUSE M, RIBEIRO V, et al. A multifractal wavelet model with application to network traffic［J］. IEEE Transactions on Information Theory, 2002, 45(3): 9921018.
［6］NORROS I. On the use of fractional Brownian motion in the theory of connectionless networks［J］. IEEE Journal on Selected Areas in Communications, 2002, 13(6): 953962.
［7］ABRY P, BARANIUK R, FLANDRIN P, et al. The multiscale nature of network traffic: discovery, analysis, and modeling［J］. IEEE Signal Processing Magazine, 2002(19): 2846.
［8］NISHIMURA S, SHINNO M. Modeling burst traffic using a MAP with a tree structure［J］. Journal of the Operations Research, 2004, 47(3): 129144. 
［9］ANDERSSON M, CAO J, KIHL M, et al. Performance modeling of an Apache web server with bursty arrival traffic［EB/OL］.［20110618］. http://lup.lub.lu.se/luur/download?func=downloadFile&recordOId=532529&fileOId=625324.
［10］KONTOVASILIS K, MITROU N. Bursty traffic modeling and efficient analysis algorithms via fluidflow models for ATM IBCN［J］. Annals of Operations Research, 1994, 49(1): 279323.
［11］LIU P, CHEUNG W, LI K. On the modeling of bursty traffic［C］//Proceedings on Computer, Communication, Control and Power Engineering. Beijing, China， 2002: 220223.
［12］JOHNSON M. Why doesn’t EM find good HMM POStaggers［C］//Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLPCoNLL). Prague, Czech Republic, 2007: 296305.
［13］FINE S, SINGER Y, TISHBY N. The hierarchical hidden Markov model: analysis and applications［J］. Machine Learning, 1998, 32(1): 4162.
［14］RABINER L. A tutorial on hidden Markov models and selected applications in speech recognition［J］. Proceedings of the IEEE, 1989, 77(2): 257286. 
Similar References:

Memo

-

Last Update: 2012-07-12

Copyright © CAAI Transactions on Intelligent Systems