Leo Budin

Genetic programming heuristics for multiple machine scheduling

In this paper we present a method for creating scheduling heuristics for parallel proportional machine scheduling environment and arbitrary performance criteria. Genetic programming is used to synthesize the priority function which, coupled with an appropriate meta-algorithm for a given environment, forms the priority scheduling heuristic. We show that the procedures derived in this way can perform similarly or better than existing algorithms. Additionally, this approach may be particularly useful for those combinations of scheduling environment and criteria for which there are no adequate scheduling algorithms.

Dynamic scheduling with genetic programming

This paper investigates the use of genetic programming in automatized synthesis of scheduling heuristics. The applied scheduling technique is priority scheduling, where the next state of the system is determined based on priority values of certain system elements. The evolved solutions are compared with existing scheduling heuristics for single machine dynamic problem and job shop scheduling with bottleneck estimation.

Solving n-Queen problem using global parallel genetic algorithm

This paper shows a way in which genetic algorithms can be used to solve the n-Queen problem. Custom chromosome representation, evaluation function and genetic operators are presented. A global parallel genetic algorithm is also demonstrated as a possible way to increase GA speed. Results are shown for several large values of n and several conclusions are drawn about solving NP problems with genetic algorithms.

Analytical prediction of performance for cache coherence protocols

In this paper, we introduce new analytical models for predicting the performance of parallel applications under various cache coherence protocol assumptions. The purpose of these models is to determine which protocols are to be used for which data blocks, and, in the case of dynamic protocols, also to determine when to change protocols. Although we focus on tightly-coupled multiprocessor systems, similar models can be derived for loosely-coupled distributed systems, such as networks of workstations. Our models are unique in that they lie between a large body of theoretical models that assume independence and a uniform distribution of memory accesses across processors, and a large body of address-trace oriented models that assume the availability of a precise characterization of interleaving behavior of memory accesses. The former are not very realistic, and the latter are not suitable for compile-time and run-time usage. In contrast, our models enable us to choose different input parameters depending on how the models will be used and depending on the needed accuracy in performance prediction. We present the models and show how the required parameters can be obtained. We assess the accuracy of our models on 15 parallel applications. For these applications, our most complete model predicts performance within a 10 percent margin when compared to a simulation of a sequentially consistent multiprocessor system. As part of this study, we also show the potential advantage of using dynamic hybrid protocols.

Undergraduate teaching of real-time scheduling algorithms by developed software tool

Theoretical foundations of scheduling in the course Real-Time Systems of the undergraduate studies program are extended by the usage of the developed visual software tool. The tool enables teaching of simulated fixed-priority scheduling algorithms. The user simply sets the simulation and algorithm properties and loads the prepared input task sets from the database. Results are obtained within the program or from external applications. The paper deals with the problem of scheduling the mixture of periodic and aperiodic tasks in a uniprocessor system. Results are displayed in the form of a flow report, table report, and diagrams of dependencies of evaluation parameters on the change of periodic and aperiodic workload. The flow report explains the algorithm behavior and the table form and dependency diagrams enable a more precise scheduling analysis and mutual comparison, respectively. Some of the evaluation parameters are the following: aperiodic response time, number of tasks with missed deadlines, utilization of resource and server tasks, resource-breakdown utilization. The software tool improves teaching of the course and the final success of students regarding relevant exam exercises.

Analytical performance evaluation of data replication based shared memory model

The proposed distributed shared memory model is based on a data replication scheme that provides an environment for a collection of processes that interact to solve a parallel programming problem. In the implementation of the scheme the authors suppose that the replicas of the shared data are present at each node and that an appropriate coherence protocol for maintaining the consistency among the replicas is applied. The performance of the distributed computation is very sensitive to the data-access behavior of the application and to the applied coherence protocol. Communication cost is regarded as an appropriate performance measure. Therefore, the authors first introduce a model characterizing the computation behavior with five workload parameters. Second, they formally describe the coherence protocols as cooperating state machines in order to evaluate their communication costs as functions of workload parameters.<<ETX>>

Static-dynamic mapping in heterogeneous computing environment

Appropriate mapping procedures make heterogeneous computing environment especially suitable for the execution of various computer-intensive applications. The paper shows a static dynamic mapping procedure relying on the multicriteria model. Model parameters describe the application, platform, mapping and user. They are divided into static and dynamic ones with three weight levels assigned to them. The described approach ensures optimal initial mapping, but also a quick response to changes in the environment, which is supported by prediction. The described approach is a simplified environment shows improvements in the execution of the test application. Further adjustments are directed towards the implementation in the computational grid.

Gradient of B-splines in volume rendering

This paper deals with surface reconstruction and the gradient reconstruction in the volume rendering. In the volume rendering procedure, reconstruction according to the discrete set of samples is required. Due to the reconstruction procedure alias artifacts in the final image, it could not be neglected. In this paper, we focus our attention on the gradient reconstruction based on the surface reconstruction. For the surface reconstruction we use the cubic B-splines and for the gradient reconstruction corresponding derivative. If the noise is present in the input data, the approximation of B-spline is used, and the interpolation B-spline is used for the input signal without the noise. The shading procedure requires normal estimation. Two approaches are used for normal estimation. The classic approach for normal estimation is central difference calculation, and we propose derivative calculation of the reconstruction B-spline function. We show that the calculation of the normal vector has important influence on the alias artifacts in the result.

Service development and application integration with public information system mediator

The software development industry is facing two important issues: enabling rapid service development and integration of existing applications. Current approaches result in custom solutions that are not generally applicable. Therefore, we propose a service development and application integration system acting as a mediator between developers, users, services, and applications. In this paper, we present the prototype of the public information system mediator MidArc. We describe the mediators system architecture, the technology it is built upon, and the process of integration of distributed applications. The mediator prototype was created and used to develop the formal automata simulator as a public service, to integrate the automata simulator service into the distance learning application, and to run the distance learning system.

Portal technology and web services as platform for process integration in virtual organizations

In this paper, we consider the construction of virtual organization utilizing Web services and portal infrastructure. Architecture for the virtual organization (VO) must provide structure and efficiency benefits, while accommodating numerous elements from multiple designers and origins. We present usage of portal technology and Web services as one way of successfully connecting individuals into virtual teams, providing them with rich working environment and enabling them to easily interact with each other, other teams and other systems, and at the same time, eliminating geographical and time obstacles