Publications
246 results found
Huang W-C, Knottenbelt WJ, 2013, Self-Adaptive Containers: Building Resource-Efficient Applications with Low Programmer Overhead, 8th International Symposium on Software Engineering for Adaptive and Self-Managing Systems (SEAMS), Publisher: IEEE, Pages: 123-132
- Author Web Link
- Cite
- Citations: 4
, 2013, Computer Performance Engineering - 10th European Workshop, EPEW 2013, Venice, Italy, September 16-17, 2013. Proceedings, Publisher: Springer
Tsimashenka I, Knottenbelt WJ, 2013, Reduction of Subtask Dispersion in Fork-Join Systems., Publisher: Springer, Pages: 325-336
, 2013, Proceedings the Sixth International Workshop on the Practical Application of Stochastic Modelling, PASM 2012, and the Eleventh International Workshop on Parallel and Distributed Methods in Verification, PDMC 2012, London, UK, September 2012, Publisher: Elsevier
Kolodziej J, Knottenbelt WJ, Khan SU, 2012, Theory and Practice of Stochastic Modelling Preface, COMPUTERS & MATHEMATICS WITH APPLICATIONS, Vol: 64, Pages: 3657-3657, ISSN: 0898-1221
Knottenbelt WJ, Spanias D, Madurska AM, 2012, A common-opponent stochastic model for predicting the outcome of professional tennis matches, COMPUTERS & MATHEMATICS WITH APPLICATIONS, Vol: 64, Pages: 3820-3827, ISSN: 0898-1221
- Author Web Link
- Cite
- Citations: 27
Malony AD, Karatza H, Knottenbelt W, et al., 2012, Topic 2: Performance prediction and evaluation, Pages: 52-53, ISSN: 0302-9743
In recent years, a range of novel methodologies and tools have been developed for the purpose of evaluation, design, and model reduction of existing and emerging parallel and distributed systems. At the same time, the coverage of the term 'performance' has constantly broadened to include reliability, robustness, energy consumption, and scalability in addition to classical performance-oriented evaluations of system functionalities. Indeed, the increasing diversification of parallel systems, from cloud computing to exascale, being fueld by technological advances, is placing greater emphasis on the methods and tools to address more comprehensive concerns. The aim of the Performance Prediction and Evaluation topic is to bring together system designers and researchers involved with the qualitative and quantitative evaluation and modeling of large-scale parallel and distributed applications and systems to focus on current critical areas of performance prediction and evaluation theory and practice. © 2012 Springer-Verlag.
Osman R, Knottenbelt WJ, 2012, Database system performance evaluation models: A survey, PERFORMANCE EVALUATION, Vol: 69, Pages: 471-493, ISSN: 0166-5316
- Author Web Link
- Cite
- Citations: 18
Tsimashenka I, Knottenbelt W, Harrison P, 2012, Controlling variability in split-merge systems, Pages: 165-177, ISSN: 0302-9743
We consider split-merge systems with heterogeneous subtask service times and limited output buffer space in which to hold completed but as yet unmerged subtasks. An important practical problem in such systems is to limit utilisation of the output buffer. This can be achieved by judiciously delaying the processing of subtasks in order to cluster subtask completion times. In this paper we present a methodology to find those deterministic subtask processing delays which minimise any given percentile of the difference in times of appearance of the first and the last subtasks in the output buffer. Technically this is achieved in three main steps: firstly, we define an expression for the distribution of the range of samples drawn from n independent heterogeneous service time distributions. This is a generalisation of the well-known order statistic result for the distribution of the range of n samples taken from the same distribution. Secondly, we extend our model to incorporate deterministic delays applied to the processing of subtasks. Finally, we present an optimisation scheme to find that vector of delays which minimises a given percentile of the range of arrival times of subtasks in the output buffer. A case study illustrates the applicability of our proposed approach. © 2012 Springer-Verlag.
Leung TLY, Knottenbelt WJ, 2012, Comparative Evaluation of Independent Private Values Distributions on Internet Auction Performance, International Journal of E-Entrepreneurship and Innovation, Vol: 3, Pages: 59-71, ISSN: 1947-8585
<p>The Independent Private Values (IPV) model is foundational for the analysis of Internet auction performance and is widely used in the study of auction behaviour. The characteristics of this model include the assumptions of privacy and independence where the value of the commodity in question is private to the individual buyers, and that different buyers do not know the values other buyers attached to the commodity. In addition, these values are drawn from a common distribution which is known to the buyers. In probabilistic terms, this essentially amounts to a series of values which are independent and identically distributed. The features and characteristics of the IPV distribution will have a significant impact on auction behaviour, and since a general stochastic analysis of their impact is analytically intractable, here auction performance is studied using an auction process simulator. Both hard close and soft close Internet auctions are studied. In addition, Vickrey auctions and auction mechanisms with multiple bid acceptance are compared and evaluated. From experimental findings, the paper establishes quantitative relationships between the different auction process parameters, deploy suitable IPV distributions to model the characteristics of different communities of bidders, provide suggestions for optimising auction performance, and recommend strategies for efficient auction design.</p>
Knottenbelt W, 2012, Message from the general chair, Proceedings - 2012 9th International Conference on Quantitative Evaluation of Systems, QEST 2012
Tsimashenka I, Knottenbelt W, 2011, Reduction of variability in split-merge systems, Pages: 101-107
We consider an optimisation problem applicable to systems that can be represented as split-merge queueing networks with a limited buffer space for processed subtasks. We assume Poisson arrivals and generally distributed service times. The proposition is to reduce variability in terms of the difference in the times of arrival of the first and last subtasks in systems where the release times of the subtasks can be controlled. This stands in contrast to the overwhelming majority of research which is focused on reduction of mean response time or percentiles of response time. We formally define our notion of variability in split-merge systems and construct an associated cost function and optimisation problem. For two case studies we use simulation to explore the optimisation landscape and to solve the associated optimisation problem.
Anastasiou N, Knottenbelt W, Marin A, 2011, Automatic synchronisation detection in Petri Net performance models derived from location tracking data, Pages: 29-41, ISSN: 0302-9743
The inference of performance models from low-level location tracking traces provides a means to gain high-level insight into customer and/or resource flow in complex systems. In this context our earlier work presented a methodology for automatically constructing Petri Net performance models from location tracking data. However, the capturing of synchronisation between service centres - the natural expression of which is one of the most fundamental advantages of Petri nets as a modelling formalism - was not explicitly supported. In this paper, we introduce mechanisms for automatically detecting and incorporating synchronisation into our existing methodology. We present a case study based on synthetic location tracking data where the derived synchronisation detection mechanism is applied. © 2011 Springer-Verlag.
Tsirimpas P, Knottenbelt WJ, 2011, SPORTSBET: A tool for the quantitative evaluation and execution of betting exchange trading strategies, Pages: 155-156
Betting exchange markets, which offer person-to-person betting, have attracted increasing interest due to their similarities with financial markets and their growing economic importance. This paper presents SPORTSBET, an event-driven tool for the quantitative evaluation of betting exchange trading strategies. It was developed to specify, execute and back-test parameterised betting strategies for a wide range of sports. SPORTSBET allows the definition of betting strategies in an extended version of the UrbiScript language as sets of concurrent processes which make use of event-calculus-like operators. Strategy performance is quantified by synchronizing multiple real time or historical data streams with a dynamic market reconstruction. © 2011 IEEE.
Leung TLY, Knottenbelt WJ, 2011, Stochastic Modelling and Optimisation of Internet Auction Processes, Electronic Notes in Theoretical Computer Science, Vol: 275, Pages: 105-121, ISSN: 1571-0661
Internet auctions are an attractive mechanism for the exchange of goods at a non-fixed price point. The operation of these auctions can be run under a variety of parameters. In this paper, we provide a theoretical analysis of fixed time forward auctions in cases where a single bid or multiple bids are accepted in a single auction. A comparison of the economic benefits and the corresponding buyer and seller surpluses between the auctions where a single bid is accepted and the auctions where multiple bids are accepted is made. These models are verified through systematic simulation experiments, based on a series of operational assumptions, which characterise the arrival rate of bids, as well as the distribution from which the private values of buyers are sampled. Decision rules for optimising surplus under different auction fee structures are also given. © 2011 Elsevier B.V.
Horng TC, Anastasiou N, Field T, et al., 2011, LocTrackJINQS: An extensible location-aware simulation tool for multiclass queueing networks, Electronic Notes in Theoretical Computer Science, Vol: 275, Pages: 93-104, ISSN: 1571-0661
This paper presents LocTrackJINQS, a flexible and extensible spatio-temporal simulation tool for systems that involve the flow and processing of customers at multiple service centres. Developed based on the multi-class queueing network simulation package JINQS, LocTrackJINQS retains the abstract model specification power of JINQS while providing additional low-level models of entity movement. Besides traditional performance metrics, LocTrackJINQS produces as output a trace of each entity's location in the system over time. It can thus be used to generate synthetic location tracking data for location-based research or applications. © 2011 Elsevier B.V.
Leung T, Knottenbelt W, 2011, The effect of private valuation on e-auction revenues, Pages: 229-242, ISSN: 1865-0929
In recent years, Internet auctions have become increasingly widespread. The Independent Private Values (IPV) model is widely used in the studying of auction behaviour and plays a fundamental role in many analyses of Internet auction performance. This model assumes privacy and independence, meaning that the private values of buyers are drawn from a common distribution, or in probabilistic terms, the series of values are independent and identically distributed. Since a general stochastic analysis is intractable, the IPV model has a significant impact on auction behaviour. In this paper, we study auction performance using an auction process simulator, considering both the hard close and soft close types of Internet auctions. From our experimental findings, we are able to establish quantitative relationships between the different auction process parameters, and also to deploy suitable IPV distributions in modelling the characteristics of different communities of bidders. © 2011 Springer-Verlag.
Franciosi F, Knottenbelt W, 2011, Data allocation strategies for the management of Quality of Service in Virtualised Storage Systems, ISSN: 2160-1968
The amount of data managed by organisations continues to grow relentlessly. Driven by the high costs of maintaining multiple local storage systems, there is a well established trend towards storage consolidation using multi-tier Virtualised Storage Systems (VSSs). At the same time, storage infrastructures are increasingly subject to stringent Quality of Service (QoS) demands. Within a VSS, it is challenging to match desired QoS with delivered QoS, considering the latter can vary dramatically both across and within tiers. Manual efforts to achieve this match require extensive and ongoing human intervention. This paper presents our work on the design and implementation of data allocation strategies in an enhanced version of the popular Linux Extended 3 Filesystem. This enhanced filesystem features support for the specification of QoS metadata while maintaining compatibility with stock kernels. We present new inode and datablock allocation strategies which seek to match the QoS attributes set by users and/or applications on files and directories with the QoS actually delivered by each of the filesystem's block groups. To create realistic test filesystems we have modified the Impressions benchmarking framework to support QoS metadata. The effectiveness of the resulting data allocation in terms of QoS matching is evaluated using a special kernel module that is capable of inspecting detailed filesystem allocation data onthe- fly. We show that our implementations of the proposed inode and datablock allocation strategies are capable of dramatically improving data placement with respect to QoS requirements when compared to the default allocators. © 2011 IEEE.
Leung TLY, Knottenbelt WJ, 2011, Consumer-to-Consumer Internet Auction Models, International Journal of Online Marketing, Vol: 1, Pages: 17-28, ISSN: 2156-1753
<p>Internet auctions have become an increasingly common method for exchanging goods and services across the world both among consumers themselves, as well as between businesses and consumers. These Internet auction mechanisms have the scope of incorporating procedures of much greater complexity and variety, and they exhibit characteristics and properties that are quite distinct from conventional auctions. In this paper, the authors provide an experimental study of the performance characteristics and operational behaviour of a number of online auction models, including the fixed time forward auctions, the Vickrey auctions, and models with soft close variable auction times. These online auction models are studied through systematic simulation experiments, based on a series of operational assumptions, which characterize the arrival rate of bids, as well as the distribution from which the private values of buyers are sampled. Suggestions for efficient online auction design and procedures for improving auction performance are given, and the behaviour of the average auction income and average auction duration are quantified and compared.</p>
Lebrecht AS, Dingle NJ, Knottenbelt WJ, 2011, Analytical and Simulation Modelling of Zoned RAID Systems, COMPUTER JOURNAL, Vol: 54, Pages: 691-707, ISSN: 0010-4620
- Author Web Link
- Cite
- Citations: 3
Knottenbelt WJ, Lebrecht AS, Dingle NJ, 2011, Performance Models of Zoned Disk Arrays, Editors: Ivanyi, Topping, Publisher: Saxe-Coburg Publications, Pages: 105-134
Guenther MC, Dingle NJ, Bradley JT, et al., 2011, Passage-time computation and aggregation strategies for large semi-Markov processes, PERFORMANCE EVALUATION, Vol: 68, Pages: 221-236, ISSN: 0166-5316
- Author Web Link
- Open Access Link
- Cite
- Citations: 4
Anastasiou N, Horng TC, Knottenbelt W, 2011, Deriving generalised stochastic Petri Net performance models from high-precision location tracking data, Pages: 91-100
Stochastic performance models have been widely used to analyse the performance and reliability of systems that involve the flow and processing of customers and/or resources with multiple service centres. However, the quality of performance analysis delivered by a model depends critically on the degree to which the model accurately represents the operations of the real system. This paper presents an automated technique which takes as input high-precision location tracking data - potentially collected from a real life system -and constructs a hierarchical Generalised Stochastic Petri Net performance model of the underlying system. We examine our method's effectiveness and accuracy through two case studies based on synthetic location tracking data. Copyright © 2011 ICST.
Al-Begain K, Fiems D, Knottenbelt W, 2010, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics): Preface, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol: 6148 LNCS, ISSN: 0302-9743
Thomas N, Bradley J, Knottenbelt W, et al., 2010, Preface, Pages: 1-4, ISSN: 1571-0661
The special issue of Electronic Notes in Theoretical Computer Science contains papers presented at the Fourth International Workshop on the Practical Application of Stochastic Modelling (PASM), held at Imperial College London, in September 2009. The papers in the issue cover a broad range of research in the area of applied stochastic modeling and involve both applications and the theory to enable practical application of modeling techniques. Wolter and Reinecke use a phase-type queuing model to investigate the performance of service-oriented systems. Harrison and Massink investigate the use of stochastic process algebra to analyze a wide range of properties of ubiquitous systems. Ciocchetta and Hillston use the Bio-PEPA process algebra to investigate a class of epidemiological model. Piazolla and Gribaudo exploit recently developed results in mean field analysis to explore the highly novel application area of television and cinema production.
Mason AM, Dingle NJ, Knottenbelt WJ, et al., 2010, Modelling infection spread using location tracking, International Journal of Healthcare Technology and Management, Vol: 11, Pages: 442-461, ISSN: 1368-2156
The precision of location tracking technology has improved greatly over the last few decades. We aim to show that by tracking the locations of individuals in a closed environment, it is now possible to record the nature and frequency of interactions between them. Further, that it is possible to use such data to predict the way in which an infection will spread throughout such a population, given parameters such as transmission and recovery rates. We accordingly present a software package that is capable of recording and then replaying location data provided by a high-precision location tracking system. The software then employs a combination of SIR modelling and the epidemiological technique of contact tracing in order to predict the spread of an infection. We use this software to conduct a number of experiments using a sample data set, and compare the SIR graphs generated from these to similar graphs generated using the traditional SIR differential equations. Copyright © 2010 Inderscience Enterprises Ltd.
, 2010, Analytical and Stochastic Modeling Techniques and Applications, 17th International Conference, ASMTA 2010, Cardiff, UK, June 14-16, 2010. Proceedings, Publisher: Springer
Kulatunga H, Knottenbelt WJ, Kadirkamanathan V, 2010, Adaptive Planning of Staffing Levels in Health Care Organisations, 2nd Congress on Electronic Healthcare for the 21st Century, Publisher: SPRINGER-VERLAG BERLIN, Pages: 88-+, ISSN: 1867-8211
Conrad J, Knottenbelt W, 2009, Message from the programme committee chairs, Proceedings - IEEE Computer Society's Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, MASCOTS, ISSN: 1526-7539
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.