Task scheduling in heterogeneous distributed systems with security and QoS requirements

Heterogeneous distributed systems are continuously being deployed, e.g., P2P and military command, control, communications, computers, and intelligence (C4I). Many of these applications trade-off quality of service (QoS) and security. In this paper, we study the problem of allocating and scheduling a set of dependent tasks with security and QoS requirements on a set of heterogeneous sites. This problem is at least as hard as the basic NP-hard problem of scheduling a directed acyclic graph (DAG) on multiple processors. In particular, we make the following contributions: 1 we formulate the problem as an optimisation problem 2 we use a branch and bound (B&B) method guided by an upper bound on total QoS value (TQV) to find the optimal solution if one exist 3 we propose a heuristic algorithm to provide a near optimal solution for the scheduling of the assigned tasks on a site to maximise the QoS provided by this site in a polynomial time 4 we evaluate the heuristic algorithm and compare it with some baseline algorithms using variations of synthetic workloads.