Memory-Resident Database Management Systems (MRDBMS) have to be
optimized for two resources: CPU cycles and memory bandwidth. To
optimize for bandwidth in mixed OLTP/OLAP scenarios, the hybrid
or Partially Decomposed Storage Model (PDSM) has been proposed.
However, in current implementations, bandwidth savings achieved
by partial decomposition come at increased CPU costs. To achieve
the aspired bandwidth savings without sacrificing CPU
efficiency, we combine partially decomposed storage with
Just-in-Time (JiT) compilation of queries, thus eliminating CPU
inefficient function calls. Since existing cost based
optimization components are not designed for JiT-compiled query
execution, we also develop a novel approach to cost modeling and
subsequent storage layout optimization. Our evaluation shows
that the JiT-based processor maintains the bandwidth savings of
previously presented hybrid query processors but outperforms
them by two orders of magnitude due to increased CPU efficiency.
%0 Conference Paper
%1 Pirk2013-xe
%A Pirk, H
%A Funke, F
%A Grund, M
%A Neumann, T
%A Leser, U
%A Manegold, S
%A Kemper, A
%A Kersten, M
%B Data Engineering (ICDE), 2013 IEEE 29th International
Conference on
%D 2013
%K All CPU_cycle Expose JiT-based_processor JiT_compilation MRDBMS Main_memory_databases OLAP OLTP PDSM bandwidth_savings cache_efficient_management cache_storage cost_based_optimization_component data_mining data_warehouses database_management_systems hybrid_decomposed_storage_model hybrid_query_processor just-in-time memory-resident_database_management_system memory_bandwidth papers3 partially_decomposed_storage_model
%P 14--25
%T CPU and cache efficient management of memory-resident
databases
%X Memory-Resident Database Management Systems (MRDBMS) have to be
optimized for two resources: CPU cycles and memory bandwidth. To
optimize for bandwidth in mixed OLTP/OLAP scenarios, the hybrid
or Partially Decomposed Storage Model (PDSM) has been proposed.
However, in current implementations, bandwidth savings achieved
by partial decomposition come at increased CPU costs. To achieve
the aspired bandwidth savings without sacrificing CPU
efficiency, we combine partially decomposed storage with
Just-in-Time (JiT) compilation of queries, thus eliminating CPU
inefficient function calls. Since existing cost based
optimization components are not designed for JiT-compiled query
execution, we also develop a novel approach to cost modeling and
subsequent storage layout optimization. Our evaluation shows
that the JiT-based processor maintains the bandwidth savings of
previously presented hybrid query processors but outperforms
them by two orders of magnitude due to increased CPU efficiency.
@inproceedings{Pirk2013-xe,
abstract = {Memory-Resident Database Management Systems (MRDBMS) have to be
optimized for two resources: CPU cycles and memory bandwidth. To
optimize for bandwidth in mixed OLTP/OLAP scenarios, the hybrid
or Partially Decomposed Storage Model (PDSM) has been proposed.
However, in current implementations, bandwidth savings achieved
by partial decomposition come at increased CPU costs. To achieve
the aspired bandwidth savings without sacrificing CPU
efficiency, we combine partially decomposed storage with
Just-in-Time (JiT) compilation of queries, thus eliminating CPU
inefficient function calls. Since existing cost based
optimization components are not designed for JiT-compiled query
execution, we also develop a novel approach to cost modeling and
subsequent storage layout optimization. Our evaluation shows
that the JiT-based processor maintains the bandwidth savings of
previously presented hybrid query processors but outperforms
them by two orders of magnitude due to increased CPU efficiency.},
added-at = {2015-04-10T18:02:47.000+0200},
author = {Pirk, H and Funke, F and Grund, M and Neumann, T and Leser, U and Manegold, S and Kemper, A and Kersten, M},
biburl = {https://www.bibsonomy.org/bibtex/2768dc0cd27f565f26a47de9f65352e74/christophv},
booktitle = {Data Engineering ({ICDE)}, 2013 {IEEE} 29th International
Conference on},
interhash = {3efec5db4fc44b72bcf11c1ce1b2e6e8},
intrahash = {768dc0cd27f565f26a47de9f65352e74},
keywords = {All CPU_cycle Expose JiT-based_processor JiT_compilation MRDBMS Main_memory_databases OLAP OLTP PDSM bandwidth_savings cache_efficient_management cache_storage cost_based_optimization_component data_mining data_warehouses database_management_systems hybrid_decomposed_storage_model hybrid_query_processor just-in-time memory-resident_database_management_system memory_bandwidth papers3 partially_decomposed_storage_model},
month = apr,
pages = {14--25},
timestamp = {2016-01-04T14:22:08.000+0100},
title = {{CPU} and cache efficient management of memory-resident
databases},
year = 2013
}