Scheduling linear chain streaming applications on heterogeneous systems with failures
Future Generation Computer Systems
| Hi-index | 0.01 |
In this paper, we study the problem of optimizing the throughput formicro-factories subject to failures. The challenge consists inmapping several tasks onto a set of machines. The originality ofour approach is the failure model for such applications in whichtasks are subject to failures rather than machines. If there is exactly one task per machine in the mapping, then we prove that the optimal solution can be computed in polynomial time.However, the problem becomes NP-hard if several tasks can beassigned to the same machine. Several polynomial time heuristics arepresented for the most realistic specialized setting, in whichtasks of a same type can be mapped onto the same machine. Experimentalresults show that the best heuristics obtain a good throughput, muchbetter than the throughput obtained with a random mapping. Moreover,we obtain a throughput close to the optimal solution in theparticular cases on which the optimal throughput can be computed.