Workflows are commonly used to model data intensive scientific analysis. As computational resource needs increase for eScience, emerging platforms like clouds present additional resource choices for scientists and policy makers. We introduce BReW, a tool enables users to make rapid, highlevel platform selection for their workflows using limited workflow knowledge. This helps make informed decisions on whether to port a workflow to a new platform. Our analysis of synthetic and real eScience workflows shows that using just total runtime length, maximum task fanout, and total data used and produced by the workflow, BReW can provide platform predictions comparable to whitebox models with detailed workflow knowledge.
%0 Conference Paper
%1 Simmhan:works:2010
%A Simmhan, Yogesh
%A Soroush, Emad
%A van Ingen, Catharine
%A Agarwal, Deb
%A Ramakrishnan, Lavanya
%B IEEE Workshop on Workflows in Support of Large-Scale Science (WORKS)
%D 2010
%K cloud, escience, msr, peer reviewed scheduling, workflow,
%P 1--10
%R 10.1109/WORKS.2010.5671857
%T BReW: Blackbox resource selection for e-Science workflows
%U http://ceng.usc.edu/~simmhan/pubs/simmhan-works-2010.pdf
%X Workflows are commonly used to model data intensive scientific analysis. As computational resource needs increase for eScience, emerging platforms like clouds present additional resource choices for scientists and policy makers. We introduce BReW, a tool enables users to make rapid, highlevel platform selection for their workflows using limited workflow knowledge. This helps make informed decisions on whether to port a workflow to a new platform. Our analysis of synthetic and real eScience workflows shows that using just total runtime length, maximum task fanout, and total data used and produced by the workflow, BReW can provide platform predictions comparable to whitebox models with detailed workflow knowledge.
@inproceedings{Simmhan:works:2010,
abstract = {Workflows are commonly used to model data intensive scientific analysis. As computational resource needs increase for eScience, emerging platforms like clouds present additional resource choices for scientists and policy makers. We introduce BReW, a tool enables users to make rapid, highlevel platform selection for their workflows using limited workflow knowledge. This helps make informed decisions on whether to port a workflow to a new platform. Our analysis of synthetic and real eScience workflows shows that using just total runtime length, maximum task fanout, and total data used and produced by the workflow, BReW can provide platform predictions comparable to whitebox models with detailed workflow knowledge.},
added-at = {2023-04-07T07:37:58.000+0200},
author = {Simmhan, Yogesh and Soroush, Emad and van Ingen, Catharine and Agarwal, Deb and Ramakrishnan, Lavanya},
biburl = {https://www.bibsonomy.org/bibtex/2e3c2d6daa094e522d280aa3520c89df5/vinayaka2000},
booktitle = {IEEE Workshop on Workflows in Support of Large-Scale Science (WORKS)},
doi = {10.1109/WORKS.2010.5671857},
interhash = {7a60ceb8ebb2cccd17095bfb1db6c645},
intrahash = {e3c2d6daa094e522d280aa3520c89df5},
keywords = {cloud, escience, msr, peer reviewed scheduling, workflow,},
month = {November},
owner = {Simmhan},
pages = {1--10},
timestamp = {2023-04-07T07:37:58.000+0200},
title = {BReW: Blackbox resource selection for e-Science workflows},
url = {http://ceng.usc.edu/~simmhan/pubs/simmhan-works-2010.pdf},
year = 2010
}