PhFit: A General Phase-Type Fitting Tool

Horváth, András; Telek, Miklós

doi:10.1007/3-540-46029-2_5

Cited by 125 publications

(75 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, a set of m individual stages are added [341,340]. Effectively, the sequence of the first n stages creates a mode that serves as the body of the distribution, and the additional individual stages are actually a hyper-exponential distribution used to construct a long tail for the distribution.…”

Section: Other Phase-type Distributionsmentioning

confidence: 99%

“…The simplest (yet quite general) approach is to create a mixture of hyper-exponential and Erlang distributions as shown in Section 3.2.6 [567,340,341]. The hyper-exponential part contributes the tail, and the Erlang part contributes the mode.…”

Section: Matching a Complete Distributionmentioning

confidence: 99%

“…Among them are a tool for fitting phase-type distributions to heavy-tailed data called PhFit (available from http://webspn.hit.bme.hu/%7Etelek/tools.htm) [341], and another called EMpht that uses the EM algorithm (available from http://home.imf.au.dk/asmus/pspapers.html) [41].…”

Section: Software Packages For Distribution Fittingmentioning

confidence: 99%

See 2 more Smart Citations

Workload Modeling for Computer Systems Performance Evaluation

Feitelson

2015

191

131

View full text Add to dashboard Cite

Reliable performance evaluations require the use of representative workloads. This is no easy task since modern computer systems and their workloads are complex, with many interrelated attributes and complicated structures. Experts often use sophisticated mathematics to analyze and describe workload models, making these models difficult for practitioners to grasp. This book aims to close this gap by emphasizing the intuition and the reasoning behind the definitions and derivations related to the workload models. It provides numerous examples from real production systems, with hundreds of graphs. Using this book, readers will be able to analyze collected workload data and clean it if necessary, derive statistical models that include skewed marginal distributions and correlations, and consider the need for generative models and feedback from the system. The descriptive statistics techniques covered are also useful for other domains.

show abstract

Section: Other Phase-type Distributionsmentioning

confidence: 99%

Section: Matching a Complete Distributionmentioning

confidence: 99%

See 1 more Smart Citation

Workload Modeling for Computer Systems Performance Evaluation

Feitelson

2015

191

131

View full text Add to dashboard Cite

show abstract

“…As an example consider (5) in which the some of the exponents of any product of the numerator (denominator) is three (two).…”

Section: A Proofs Of Theorem 9 and 10mentioning

confidence: 99%

“…There are two main categories of phase type fitting algorithms: numerical optimization [1,5] and procedures that make use of explicit expressions. Up to now explicit expressions for matching only the first three moments of acyclic phase type distributions (APH) have been known [9,7,8,2].…”

Section: Introductionmentioning

confidence: 99%

Matching More Than Three Moments with Acyclic Phase Type Distributions

Horváth

Telek

2007

Stochastic Models

Self Cite

View full text Add to dashboard Cite

This paper presents a matching procedure for generating an acyclic phase type distribution of order N given the first 2N − 1 moments, if they are feasible. The matching procedure uses an iterative approach and, theoretically, it can be applied to match an arbitrary number of moments.The first step of the iterative procedure contains the solution of an equation of order N and the order is decreased by one in each consecutive step. Apart of these equations the procedure makes use of explicit expressions. The practical applicability of the proposed procedure is limited by the numerical accuracy of the solution of these equations and the complexity of the involved expressions. We present examples for matching more than 10 moments with acyclic phase type distributions.

show abstract