Bharat Sukhwani , IBM Research, Yorktown Hts.
Hong Min , IBM Research, Yorktown Hts.
Mathew Thoennes , IBM Research, Yorktown Hts.
Parijat Dube , IBM Research, Yorktown Hts.
Bernard Brezzo , IBM Research, Yorktown Hts.
Sameh Asaad , IBM Research, Yorktown Hts.
Donna Dillenberger , IBM Research, Yorktown Hts.
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MM.2013.107
Businesses are experiencing a growing need for performing real-time analytics on ever increasing enterprise data. While providing useful business insights and improved market responsiveness, analytics also adds computational burden on traditional online transaction processing (OLTP) systems and may adversely affect the performance of OLTP workloads. We present a highly-pipelined, high throughput query processing engine on Field Programmable Gate Array (FPGA) to offload expensive queries for database analytics. The proposed solution provides a mechanism for a database management system (DBMS) to seamlessly harness the FPGA accelerator without requiring any changes in the application or the existing data layout. Our system, which uses an off-the-shelf server and a PCIe-attached FPGA card and is integrated into a commercial DBMS platform, achieves an order of magnitude speedup on a variety of real-life queries.
Bharat Sukhwani, Hong Min, Mathew Thoennes, Parijat Dube, Bernard Brezzo, Sameh Asaad, Donna Dillenberger, "Database Analytics: A Reconfigurable Computing Approach", IEEE Micro, , no. 1, pp. 1, PrePrints PrePrints, doi:10.1109/MM.2013.107