Other formats:
BibTeX
LaTeX
RIS
@inproceedings{1313981,
author = {Makatun, Dzmitry and Lauret, Jerome and Rudová, Hana and Šumbera, Michal},
address = {Prague, Czech Republic},
booktitle = {Journal of Physics: Conference Series, vol. 608},
doi = {http://dx.doi.org/10.1088/1742-6596/608/1/012028A},
keywords = {planning; constraint programming; distributed computational resources; STAR experiment},
howpublished = {tištěná verze "print"},
language = {eng},
location = {Prague, Czech Republic},
pages = {1-6},
publisher = {Institute of Physics Publishing},
title = {Planning for distributed workflows: constraint-based coscheduling of computational jobs and data placement in distributed environments},
url = {http://dx.doi.org/10.1088/1742-6596/608/1/012028A},
year = {2015}
}
TY - JOUR
ID - 1313981
AU - Makatun, Dzmitry - Lauret, Jerome - Rudová, Hana - Šumbera, Michal
PY - 2015
TI - Planning for distributed workflows: constraint-based coscheduling of computational jobs and data placement in distributed environments
PB - Institute of Physics Publishing
CY - Prague, Czech Republic
KW - planning
KW - constraint programming
KW - distributed computational resources
KW - STAR experiment
UR - http://dx.doi.org/10.1088/1742-6596/608/1/012028A
L2 - http://dx.doi.org/10.1088/1742-6596/608/1/012028A
N2 - When running data intensive applications on distributed computational resources long I/O overheads may be observed as access to remotely stored data is performed. Latencies and bandwidth can become the major limiting factor for the overall computation performance and can reduce the CPU/WallTime ratio to excessive IO wait. Reusing the knowledge of our previous research, we propose a constraint programming based planner that schedules computational jobs and data placements (transfers) in a distributed environment in order to optimize resource utilization and reduce the overall processing completion time. The optimization is achieved by ensuring that none of the resources (network links, data storages and CPUs) are oversaturated at any moment of time and either (a) that the data is pre-placed at the site where the job runs or (b) that the jobs are scheduled where the data is already present. Such an approach eliminates the idle CPU cycles occurring when the job is waiting for the I/O from a remote site and would have wide application in the community. Our planner was evaluated and simulated based on data extracted from log files of batch and data management systems of the STAR experiment. The results of evaluation and estimation of performance improvements are discussed in this paper.
ER -
MAKATUN, Dzmitry, Jerome LAURET, Hana RUDOVÁ and Michal ŠUMBERA. Planning for distributed workflows: constraint-based coscheduling of computational jobs and data placement in distributed environments. In \textit{Journal of Physics: Conference Series, vol. 608}. Prague, Czech Republic: Institute of Physics Publishing, 2015, p.~1-6. ISSN~1742-6588. Available from: https://dx.doi.org/10.1088/1742-6596/608/1/012028A.
|
