Heuristic-Based Architecture Generation for Complex Computer System Optimisation

Maxwell, Christopher D.; Parakhine, Artem; Leaney, John; OiNeill, T.; Denford, M.

doi:10.1109/ecbs.2005.43

Cited by 7 publications

(7 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In previous work we have highlighted that the use of heuristics could be a suitable mechanism for storing the design expertise associated with computer-based system architecture and this expertise can be used to generate new designs [8,10].…”

Section: Figure 2 Architectural Optimisation Frameworkmentioning

confidence: 99%

“…The goal being to find designs which have the best compromise between the system qualities, represented by the top-right corner of the figure. [10,8].…”

Section: Introductionmentioning

confidence: 97%

“…One approach that shows promise, however, is design space exploration. Design space exploration attempts to optimise the architecture of a computerbased system by generating many potential architectural designs that meet the requirements of the system [10]. These designs are then compared to see which design has the best combination of non-functional system properties.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Utilising Abstract Matching to Preserve the Nature of Heuristics in Design Optimisation

Maxwell

Leaney

O'Neill³

2008

15th Annual IEEE International Conference and Workshop on the Engineering of Computer Based Systems (Ecbs 2008)

View full text Add to dashboard Cite

Design space exploration, the generation of alternate designs to identify working designs with varying system properties, has the potential to provide a basis for the optimisation of computer-based system architectures. To utilise design space exploration for this purpose requires that an effective mechanism exist for the storage and application of potential design changes. Heuristics have shown some promise in this area due to their ability to capture expert design knowledge and their flexibility across multiple domains. Heuristics are also especially attractive as change descriptions as they can capture changes that operate across a large spectrum of change detail, from the very detailed to the very abstract. Heuristics are, however, at their most powerful, and their most useful, when they are specified in an abstract manner. This presents a challenge in the formal application of heuristics in capturing design knowledge. Formally describing heuristics inclines their specification of change to be in a more detailed, more concrete, state than an abstract one. This occurs because architectural models tend be both domain specific and are often described at a more concrete level than the level at which the heuristic is described. This has the potential to greatly reduce the effectiveness of heuristics. In this paper we propose that by providing an abstract match method heuristics may be specified in an abstract manner and still be applied to a detailed formal model, thereby eliminating this problem.

show abstract

Section: Figure 2 Architectural Optimisation Frameworkmentioning

confidence: 99%

“…The goal being to find designs which have the best compromise between the system qualities, represented by the top-right corner of the figure. [10,8].…”

Section: Introductionmentioning

confidence: 97%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Utilising Abstract Matching to Preserve the Nature of Heuristics in Design Optimisation

Maxwell

Leaney

O'Neill³

2008

15th Annual IEEE International Conference and Workshop on the Engineering of Computer Based Systems (Ecbs 2008)

View full text Add to dashboard Cite

show abstract

“…After an initial proposal and exploration of the problem space [18] we have developed an optimisation framework for optimising non-functional qualities in computer-based systems. Figure 1 gives a highlevel view of this framework.…”

Section: Framework Overviewmentioning

confidence: 99%

Formal Architecture Transformation Using Heuristics

Maxwell

O'Neill

Leaney

2007

14th Annual IEEE International Conference and Workshops on the Engineering of Computer-Based Systems (ECBS'07)

View full text Add to dashboard Cite

Heuristics have long been a popular and effective mechanism for capturing the knowledge of experts. In recent times, however, the more common use of heuristics has been as a means for communicating ideas at an abstract level, with little consideration to their potential as a structured approach to design improvement. With this paper we present the issues surrounding, and a structured method for, formally capturing architectural change embodied within heuristics. We demonstrate how through the application of graph theory, category theory and predicate calculus we can capture change within a heuristic and then use it to achieve formal heuristic-based transformation of a realworld system. By capturing heuristics in the structured and formal manner discussed in this paper we present ourselves with the opportunity to create a practical and reliable heuristic-based architecture transformation system. This is done within the wider context of achieving a process for optimising the non-functional qualities of a system architecture through design transformation.

show abstract

“…The ultimate aim of the process is to find a solution that represents an optimal compromise on competing system qualities. To achieve a positive outcome the process requires priming, which includes obtaining a set of potential design changes encoded as heuristics (Maxwell et al, 2005), providing a baseline design, and a set of goals and constraints. Effectively, the goals and constraints must contain the following: the expression of desired non-functional qualities that should be present in the solution and the minimal levels to which these qualities must be achieved for a given solution to be considered valid.…”

Section: Heuristic-based Optimisation Frameworkmentioning

confidence: 99%

Guiding Complex Design Optimisation Using Bayesian Networks

Leaney¹,

Parakhine²

2010

Bayesian Network

View full text Add to dashboard Cite

The process of design optimisation is a special form of design synthesis characterised by a high level of specificity with respect to its goals. As such it relies on the following supporting elements: measurable goals -clear and quantifiable statement of goals to enable evaluation of candidate designs.design variability -the system design must present amenable elements. In the context of system design the change may be made at various levels: topology of component arrangement, interchangeable components, configuration within a component.constraints -restrictions, such as budgeted cost and time, provide an essential definition of design feasibility which effectively reduces the search space to a manageable size. 7www.intechopen.com Bayesian Network 110Currently, all prevalent approaches to architectural system design can be separated into two groups based on their main artefacts: quality-driven and model-driven (Parakhine, 2009, p. 22). The quality-driven design synthesis process focuses on extensive analysis of qualities and possible trade-offs. It necessitates creation of goal descriptions expressed partially in terms of choices required to attain them, which in turn contributes towards the evaluation of the candidate architectures. On the other hand, the model-driven methodology concentrates on representing the system design from the point of view of a specific domain (e.g. security) with consequent modification through application of well-defined domain-specific changes. Although the two groups operate on different artefacts, both of them fit into the same general structure of an architectural optimisation process shown in Figure 1. Fig. 1. A generic process of system optimisation at an architectural level.In its simplest terms, the process of architectural optimisation can be characterised as one based on iterative derivation and evaluation aimed at satisfaction of multiple goals. The fields of computer science and system design offer a wealth of frameworks exhibiting features and capabilities of the optimisation process shown in Figure 1. Table 1 details how the requirements for facilitation of architectural optimisation have been addressed by some of the various frameworks built specifically for that purpose or built to solve general design problems in a way which makes optimisation possible. The described optimisation approaches adopt various methods and techniques to attain their goals and produce the outcomes. Those relevant to the system design process are listed below. The presented features and techniques are grouped by the element of the optimisation process (Figure 1) whose function they were used to fulfill:Genetic Algorithms -indicates that the framework performs evolutionary search that relies on encoding of system design in the form appropriate for breeding operators.Meta-Heuristics -shows that the framework was created around a specific optimisation heuristic, such as simulated annealing or a tabu-search.Function Analysis -identifies frameworks that employ a mathematical function analysis to ...

show abstract

Heuristic-Based Architecture Generation for Complex Computer System Optimisation

Cited by 7 publications

References 14 publications

Utilising Abstract Matching to Preserve the Nature of Heuristics in Design Optimisation

Utilising Abstract Matching to Preserve the Nature of Heuristics in Design Optimisation

Formal Architecture Transformation Using Heuristics

Guiding Complex Design Optimisation Using Bayesian Networks

Contact Info

Product

Resources

About