Throughput Optimization for Pipeline Workflow Scheduling with Setup Times
We tackle pipeline workflow applications that are executed on a distributed platform with setup times. Several computation stages are interconnected as a linear application graph, and each stage holds a buffer of limited size where intermediate results are stored and a processor setup time occurs when passing from one stage to another. In this paper, we focus on interval mappings (consecutive stages mapped on a same processor), and the objective is the throughput optimization. Even when neglecting setup times, the problem is NP-hard on heterogeneous platforms and we therefore restrict to homogeneous resources. We provide an optimal algorithm for constellations with identical buffer capacities. When buffer sizes are not fixed, we deal with the problem of allocating the buffers in shared memory and present a b/(b + 1)-approximation algorithm.
Unable to display preview. Download preview PDF.
- 3.Benoit, A., Coqblin, M., Nicod, J.M., Philippe, L., Rehn-Sonigo, V.: Throughput optimization for pipeline workflow scheduling with setup times. Research Report 7886, INRIA (2012), http://graal.ens-lyon.fr/~abenoit/papers/RR-7886.pdf
- 5.Bhat, P., Raghavendra, C., Prasanna, V.: Efficient collective communication in distributed heterogeneous systems. In: 19th ICDCS 1999, pp. 15–24 (1999)Google Scholar
- 9.Li, L., Qiao, F.: Aco-based scheduling for a single batch processing machine in semiconductor manufacturing. In: IEEE Int. CASE 2008, pp. 85–90 (2008)Google Scholar
- 10.Li, L., Qiao, F., Wu, Q.: Aco-based scheduling of parallel batch processing machines to minimize the total weighted tardiness. In: Int. CASE 2009, pp. 280–285 (2009)Google Scholar
- 14.Subhlok, J., Vondran, G.: Optimal latency-throughput tradeoffs for data parallel pipelines. In: Proceedings of the Eighth Annual ACM Symposium on Parallel Algorithms and Architectures, p. 71. ACM (1996)Google Scholar
- 15.Zhang, M., Goldberg, K.: Calibration of wafer handling robots: A fixturing approach. In: IEEE Int. CASE 2007, pp. 255–260 (2007)Google Scholar