Scipediacontent: Scipediacontent moved page Draft Content 768303027 to Rehn-Sonigo et al 2014a

2021-01-28T18:07:08Z

Scipediacontent moved page Draft Content 768303027 to Rehn-Sonigo et al 2014a

Scipediacontent: Created page with " == Abstract == Mapping linear workflow applications onto a set of homogeneous processors can be optimally solved in polynomial time for the throughput objective with fewer p..."

2021-01-28T18:07:04Z

Created page with " == Abstract == Mapping linear workflow applications onto a set of homogeneous processors can be optimally solved in polynomial time for the throughput objective with fewer p..."

New page

== Abstract ==

Mapping linear workflow applications onto a set of homogeneous processors can be optimally solved in polynomial time for the throughput objective with fewer processors than stages. This result even holds true, when setup times occur in the execution and homogeneous buffers are available for the storage of intermediate results. In this kind of applications, 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 tackle the problem where the buffer sizes are not given beforehand and have to be fixed before the execution to maximize the throughput within each processor. The goal of this work is to minimize the cost induced by the setup times allocating buffers with proportional sizes of each other. We present a closed formula to compute the optimal buffer allocation in the case of non-decreasing setup costs in the linear application. For the case of unsorted setup times, we provide competitive heuristics that are validated via extensive simulation.

== Original document ==

The different versions of the original document can be found in:

* [http://graal.ens-lyon.fr/~abenoit/papers/apdcm14.pdf http://graal.ens-lyon.fr/~abenoit/papers/apdcm14.pdf]

* [http://xplorestaging.ieee.org/ielx7/6967893/6969353/06969447.pdf?arnumber=6969447 http://xplorestaging.ieee.org/ielx7/6967893/6969353/06969447.pdf?arnumber=6969447],
: [http://dx.doi.org/10.1109/ipdpsw.2014.77 http://dx.doi.org/10.1109/ipdpsw.2014.77]

* [https://dblp.uni-trier.de/db/conf/ipps/ipdps2014w.html#BenoitNR14 https://dblp.uni-trier.de/db/conf/ipps/ipdps2014w.html#BenoitNR14],
: [http://scheduling2014.sciencesconf.org/conference/scheduling2014/pages/Veronika_sonigo_presentation_lyon.pdf http://scheduling2014.sciencesconf.org/conference/scheduling2014/pages/Veronika_sonigo_presentation_lyon.pdf],
: [https://academic.microsoft.com/#/detail/1985540689 https://academic.microsoft.com/#/detail/1985540689]

← Older revision	Revision as of 18:07, 28 January 2021
(No difference)

Rehn-Sonigo et al 2014a - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 768303027 to Rehn-Sonigo et al 2014a

Scipediacontent: Created page with " == Abstract == Mapping linear workflow applications onto a set of homogeneous processors can be optimally solved in polynomial time for the throughput objective with fewer p..."