Scheduling divisible loads on heterogeneous desktop systems with limited memory
Euro-Par'11 Proceedings of the 2011 international conference on Parallel Processing
| Hi-index | 0.00 |
During the last decade, using parallel and distributed system has been more general. In these systems a huge size of data or computation is distributed between many systems for getting better performance. Dividing data is one of the challenges in this type of system. Because efficiency of system has direct relation to distributed data. One of proposed method for managing data distribution called Divisible Load Theory. This method is based on this actuality that data in some applications can be divided arbitrarily to some independent parts which each of them can be processed independently from others. In this paper we present a survey on DLT. This article will discuss about some basic concepts for understanding this method. Then some reasons for using it and basic concept and notation used in this method, will be investigated. We will solve an example and some of application of this method will be mentioned.