To share or not to share? Export

@inproceedings{citeulike:3171651, abstract = {Intuitively, aggressive work sharing among concurrent queries in a database system should always improve performance by eliminating redundant computation or data accesses. We show that, contrary to common intuition, this is not always the case in practice, especially in the highly parallel world of chip multiprocessors. As the number of cores in the system increases, a trade-off appears between exploiting work sharing opportunities and the available parallelism. To resolve the trade-off, we develop an analytical approach that predicts the effect of work sharing in multi-core systems. Database systems can use the model to determine, statically or at runtime, whether work sharing is beneficial and apply it only when appropriate.}, author = {Johnson, Ryan and Harizopoulos, Stavros and Hardavellas, Nikos and Sabirli, Kivanc and Pandis, Ippokratis and Ailamaki, Anastasia and Mancheril, Naju G. and Falsafi, Babak}, booktitle = {VLDB '07: Proceedings of the 33rd international conference on Very large data bases}, citeulike-article-id = {3171651}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1325851.1325894}, isbn = {978-1-59593-649-3}, keywords = {2007, cmu, database, shore, vldb}, location = {Vienna, Austria}, pages = {351--362}, posted-at = {2008-08-29 12:04:54}, priority = {2}, publisher = {VLDB Endowment}, title = {To share or not to share?}, url = {http://portal.acm.org/citation.cfm?id=1325851.1325894}, year = {2007} }

TY - CONF ID - citeulike:3171651 L3 - citeulike-article-id:3171651 N2 - Intuitively, aggressive work sharing among concurrent queries in a database system should always improve performance by eliminating redundant computation or data accesses. We show that, contrary to common intuition, this is not always the case in practice, especially in the highly parallel world of chip multiprocessors. As the number of cores in the system increases, a trade-off appears between exploiting work sharing opportunities and the available parallelism. To resolve the trade-off, we develop an analytical approach that predicts the effect of work sharing in multi-core systems. Database systems can use the model to determine, statically or at runtime, whether work sharing is beneficial and apply it only when appropriate. CY - Vienna, Austria SN - 978-1-59593-649-3 EP - 362 TI - To share or not to share? PB - VLDB Endowment T2 - VLDB '07: Proceedings of the 33rd international conference on Very large data bases SP - 351 KW - 2007 KW - cmu KW - database KW - shore KW - vldb AU - Johnson, Ryan AU - Harizopoulos, Stavros AU - Hardavellas, Nikos AU - Sabirli, Kivanc AU - Pandis, Ippokratis AU - Ailamaki, Anastasia AU - Mancheril, Naju AU - Falsafi, Babak PY - 2007/// UR - http://portal.acm.org/citation.cfm?id=1325851.1325894 ER -