Wyniki wyszukiwania - BazTech

1

Particle Swarm Optimization based Sequential and Parallel Tasks Scheduling Model for Heterogeneous Multiprocessor Systems

Kumar P. R., Babu P., Palani S.

Fundamenta Informaticae

|

2015

|

Vol. 139, nr 1

43--65

EN

Multiprocessors have emerged as a powerful computing means for running real-time applications, especially where a uniprocessor system would not be sufficient enough to execute all the tasks. The high performance and reliability of multiprocessors have made them a powerful computing resource. Such computing environment requires an efficient algorithm to determine when and on which processor a given task should be executed. In multiprocessor systems, an efficient scheduling of sequential and parallel tasks onto the processors is known to be NP- Hard problem. In this paper, the problem of minimizing schedule length with energy consumption constraint and the problem of minimizing energy consumption with schedule length constraint on homogeneous and heterogeneous multiprocessor computers through independent sequential and parallel tasks are proposed. These problems emphasize the tradeoff between power and performance and are defined such that the power-performance product is optimized by fixing one factor and minimizing the other and vice versa. The performance of the proposed algorithm with optimal solution is validated using Particle Swarm Optimization (PSO) The PSO algorithm achieves 47.5% and 32% of power savings for scheduling sequential and parallel tasks to the processors respectively and also 45.5% of energy saving are achieved for scheduling both sequential or parallel tasks to the processors.

2

Efficient data generation for the testing of real-time multiprocessor scheduling algorithms

Shehzad M. N, Déplanche A. M, Trinquet Y, Farooq U

Przegląd Elektrotechniczny

|

2014

|

R. 90, nr 9

136-139

EN

In conducting the performance evaluation tests of real-time multiprocessor scheduling algorithms, synthetic input data is mandatory for obtaining reliable results and to draw strong conclusions. The results of the statistical evaluation highly depend on the data chosen for the experimentation. The data generation process is required to be efficient while the resultant data should comply with the user requirements. This article discusses two established data generation techniques used in the literature, explains their advantages and disadvantages, and finally proposes few extensions in one of the techniques, which is considered quite efficient, to be made compatible with discrete time simulator.

PL

W artykule przedyskutowano dwie techniki generacji danych w multiprocesorze czasu rzeczywistego. Zaproponowano modyfikacje algorytmu Stafforda.

3

Rozszerzenia funkcjonalne arbitrażu magistrali VME

Wiśniewski B., Szecówka-Wiśniewska B.

Elektronika : konstrukcje, technologie, zastosowania

|

2011

|

Vol. 52, nr 10

113-115

PL

Omówiono zasady arbitrażu magistrali VME i obowiązujące standardy. Przedstawiono zasady rozszerzeń funkcjonalnych arbitrażu z uwzględnieniem specyfiki laboratoryjnej. Zaproponowaną rozbudowę funkcjonalną zaprezentowano w postaci sieci działania, w odniesieniu do zminimalizowanego rozwiązania układowego.

EN

VME bus arbiter and obligatory standards are discussed. Principles of functional arbiter's expansions, are presented. Proposed functional extension in form of flow chart is shown, in relation to minimized circuit solution.

4

Parallelization of Diacoptic Methods for Multiprocessor Computing Systems

Stakhiv P., Rendzinyak S., Krupskyy B.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2003

|

Vol. 51, nr 4

381-394

EN

Parallelization process features of dynamic regimes calculation of electronic circuits on multiprocessor computing systems are considered.

5

On-line algorithms for multiprocessor task scheduling with ready times

Caramia M., Dell'Olmo P., Iovanella A.

Foundations of Computing and Decision Sciences

|

2001

|

Vol. 26, No. 3

197-214

EN

In this paper we deal with multiprocessor task scheduling with ready times and prespecified processor allocation. In the studied problem tasks are not initially all available in the scheduler and can be executed only by a given ready time. Moreover, a task can be examined in order to be processed only when it enters the scheduler. For this class of problems we developed some algorithms which schedule tasks in the attempt to minimize the makespan. We provide experiments on various scenarios, computing also the mean flow time spent by each task in the system.