Abstract

AbstractWe consider a set of n processors p1, P2,…,pn which communicate via a network and a modular program consisting of m tasks t1, t2,…, tm. Tasks can be assigned to any processor and some of them exchange data.The network configuration can be represented by a graph Gp = (Vp, Ep); the set of vertices Vp corresponds to the set of processors and two vertices are linked by an edge if the corresponding processors can communicate. Together with the set of tasks, we have a directed or undirected graph Gt = (Vt, Et), called the communication graph, whose vertices are the tasks of the program and such that there is an arc (or an edge) between two vertices if the corresponding tasks communicate.We study here the problem of determining an optimal assignment of the tasks which satisfies some constraints. We consider the three main classes of models studied in the bibliography. In the first class, the objective is to make the best use of resources and the objective function represents the global processing cost an...

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call