A Novel Approach for Phase-Type Fitting with the EM Algorithm

Authors:
Axel Thummler;Peter Buchholz;Miklos Telek
Affiliations:
-;IEEE;-
Venue:
IEEE Transactions on Dependable and Secure Computing
Year:
2006

Citing 5
Cited 22

Probability and statistics with reliability, queuing and computer science applications

Probability and statistics with reliability, queuing and computer science applications
Hyper-Erlang distribution model and its application in wireless mobille networks

Wireless Networks - Special issue: Design and modeling in mobile and wireless systsems
Markovian Modeling of Real Data Traffic: Heuristic Phase Type and MAP Fitting of Heavy Tailed and Fractal Like Samples

Performance Evaluation of Complex Systems: Techniques and Tools, Performance 2002, Tutorial Lectures
PhFit: A General Phase-Type Fitting Tool

TOOLS '02 Proceedings of the 12th International Conference on Computer Performance Evaluation, Modelling Techniques and Tools
Fitting world-wide web request traces with the EM-algorithm

Performance Evaluation - Special issue: Internet performance and control of network systems

Response-time measurements using the sun Java adventure builder

Proceedings of the 1st international workshop on Quality of service-oriented software systems
Stochastic Models for Dependable Services

Electronic Notes in Theoretical Computer Science (ENTCS)
Accuracy of link status detection in wireless multi-hop networks

Proceedings of the 13th ACM international conference on Modeling, analysis, and simulation of wireless and mobile systems
A novel framework for handoff analysis under generalized session and mobility statistics

Proceedings of the 13th ACM international conference on Modeling, analysis, and simulation of wireless and mobile systems
Multi-class Markovian arrival processes and their parameter fitting

Performance Evaluation
Performance evaluation of a single node with general arrivals and service

ASMTA'11 Proceedings of the 18th international conference on Analytical and stochastic modeling techniques and applications
Correlated phase-type distributed random numbers as input models for simulations

Performance Evaluation
Deriving generalised stochastic Petri net performance models from high-precision location tracking data

Proceedings of the 5th International ICST Conference on Performance Evaluation Methodologies and Tools
An empirical comparison of MAP fitting algorithms

MMB&DFT'10 Proceedings of the 15th international GI/ITG conference on Measurement, Modelling, and Evaluation of Computing Systems and Dependability and Fault Tolerance
Reducing the cost of generating APH-Distributed random numbers

MMB&DFT'10 Proceedings of the 15th international GI/ITG conference on Measurement, Modelling, and Evaluation of Computing Systems and Dependability and Fault Tolerance
ProFiDo – a toolkit for fitting input models

MMB&DFT'10 Proceedings of the 15th international GI/ITG conference on Measurement, Modelling, and Evaluation of Computing Systems and Dependability and Fault Tolerance
Automatic synchronisation detection in petri net performance models derived from location tracking data

EPEW'11 Proceedings of the 8th European conference on Computer Performance Engineering
Phase-type distributions for realistic modelling in discrete-event simulation

Proceedings of the 5th International ICST Conference on Simulation Tools and Techniques
Cluster-based fitting of phase-type distributions to empirical data

Computers & Mathematics with Applications
Traffic modeling with a combination of phase-type distributions and ARMA processes

Proceedings of the Winter Simulation Conference
An efficient MCMC algorithm for continuous ph distributions

Proceedings of the Winter Simulation Conference
Multiple class G-networks with restart

Proceedings of the 4th ACM/SPEC International Conference on Performance Engineering
Deriving coloured generalised stochastic petri net performance models from high-precision location tracking data

Proceedings of the 4th ACM/SPEC International Conference on Performance Engineering
Retrial queuing system with Markovian arrival flow and phase-type service time distribution

Computers and Industrial Engineering
PEPERCORN: inferring performance models from location tracking data

QEST'13 Proceedings of the 10th international conference on Quantitative Evaluation of Systems
Fault-Impact models based on delay and packet loss for IEEE 802.11g

QEST'13 Proceedings of the 10th international conference on Quantitative Evaluation of Systems
Approximate aggregation of Markovian models using alternating least squares

Performance Evaluation

Quantified Score

Hi-index	0.01

Visualization

Abstract

The representation of general distributions or measured data by phase-type distributions is an important and nontrivial task in analytical modeling. Although a large number of different methods for fitting parameters of phase-type distributions to data traces exist, many approaches lack efficiency and numerical stability. In this paper, a novel approach is presented that fits a restricted class of phase-type distributions, namely, mixtures of Erlang distributions, to trace data. For the parameter fitting, an algorithm of the expectation maximization type is developed. This paper shows that these choices result in a very efficient and numerically stable approach which yields phase-type approximations for a wide range of data traces that are as good or better than approximations computed with other less efficient and less stable fitting methods. To illustrate the effectiveness of the proposed fitting algorithm, we present comparative results for our approach and two other methods using six benchmark traces and two real traffic traces as well as quantitative results from queueing analysis.