الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 103 |  |  |
| | | from | 103 | | |
Abstract
There has been a great increase in the popularity of heterogeneous computing systems that provide various computing resources and multiplex many users on the same physical infrastructure. The heterogeneous computing is a special style of computing. It has many advantages along with some crucial issues to be resolved in order to improve the performance of its environment. These issues are related with resource management, fault tolerance and different security issues for many applications. The purpose of this research is to handle service flow management problem that are running on the heterogeneous resource. The management means how to allocate and schedule the incoming services on available candidate in a way that help in reducing the completion time and increasing user satisfaction.
The problem of service flow management is handled from various aspects, using genetic technique. Firstly a model for service flow management is proposed. This model consist of four modules that models the service flow needs in heterogeneous computing.
Secondly, the problem of service flow management is formulated as a single-objective problem that aims to reduce the consumed cost and a multi- objective problem that aims to optimize the security level and minimize the cost and completion time. After that a two service flow mapping algorithm based genetic to solve the formulated problems are proposed.
The proposed algorithms give a robust search technique that allow a soft cost solution to be derived from a huge search space of solutions by inheriting the evolution concepts. They also focuses on the improvement of the security level, optimizes the execution time objective to meet the deadline constraint and minimizes the execution cost according to the budget in heterogeneous computing. The obtained results from the applied experiments prove that genetic can save more than fifteen percent from the cost and also outperforms the compared algorithms.