Adaptive run-time prediction in heterogeneous environments Export

@inproceedings{citeulike:4921079, abstract = {In this article we describe an approach for the prediction of the run-time of jobs in heterogeneous environments that applies a meta-prediction algorithm working in multiple phases. For an efficient utilization of hardware resources, it is necessary to support schedulers with detailed information about the jobs that are going to be dispatched. One technique is to provide accurate forecasts of application run-times. A large number of current approaches focus on limited sets of prediction techniques (often linear ones) whereas most times only one is deployed on a dataset, ignoring the different characteristics that are "denoted" inherently by the data and therefore are not obvious. For that reason we are proposing an adaptive system for run-time prediction that offers a large set of different and differently parameterized predictors respectively, whereas only the appropriate prediction techniques for specifically filtered clusters of jobs are executed.}, address = {New York, NY, USA}, author = {Glasner, Christian and Volkert, Jens}, booktitle = {HPDC '09: Proceedings of the 18th ACM international symposium on High performance distributed computing}, citeulike-article-id = {4921079}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1551609.1551622}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1551609.1551622}, doi = {10.1145/1551609.1551622}, isbn = {978-1-60558-587-1}, keywords = {algorithms, concurrency, distributed, theory}, location = {Garching, Germany}, pages = {61--62}, posted-at = {2009-06-22 08:59:29}, priority = {2}, publisher = {ACM}, title = {Adaptive run-time prediction in heterogeneous environments}, url = {http://dx.doi.org/10.1145/1551609.1551622}, year = {2009} }

TY - CONF ID - citeulike:4921079 L3 - citeulike-article-id:4921079 N2 - In this article we describe an approach for the prediction of the run-time of jobs in heterogeneous environments that applies a meta-prediction algorithm working in multiple phases. For an efficient utilization of hardware resources, it is necessary to support schedulers with detailed information about the jobs that are going to be dispatched. One technique is to provide accurate forecasts of application run-times. A large number of current approaches focus on limited sets of prediction techniques (often linear ones) whereas most times only one is deployed on a dataset, ignoring the different characteristics that are "denoted" inherently by the data and therefore are not obvious. For that reason we are proposing an adaptive system for run-time prediction that offers a large set of different and differently parameterized predictors respectively, whereas only the appropriate prediction techniques for specifically filtered clusters of jobs are executed. CY - Garching, Germany SN - 978-1-60558-587-1 AD - New York, NY, USA EP - 62 TI - Adaptive run-time prediction in heterogeneous environments PB - ACM T2 - HPDC '09: Proceedings of the 18th ACM international symposium on High performance distributed computing SP - 61 KW - algorithms KW - concurrency KW - distributed KW - theory AU - Glasner, Christian AU - Volkert, Jens PY - 2009/// UR - http://dx.doi.org/10.1145/1551609.1551622 ER -