Accelerating Join Operation for Relational Databases with FPGAs

被引：41

作者：

Halstead, Robert J. ^{[1
]}

Sukhwani, Bharat ^{[2
]}

Min, Hong ^{[2
]}

Thoennes, Mathew ^{[2
]}

Dube, Parijat ^{[2
]}

Asaad, Sameh ^{[2
]}

Iyer, Balakrishna ^{[3
]}

机构：

[1] Univ Calif Riverside, Dept Comp Sci, Riverside, CA 92521 USA

[2] IBM T J Watson Res Ctr, Yorktown Hts, NY USA

[3] IBM Silicon Valley Lab, San Jose, CA USA

来源：

2013 IEEE 21ST ANNUAL INTERNATIONAL SYMPOSIUM ON FIELD-PROGRAMMABLE CUSTOM COMPUTING MACHINES (FCCM) | 2013年

关键词：

Relational database; analytics; FPGA; hardware aceleration; database join;

D O I：

10.1109/FCCM.2013.17

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this paper, we investigate the use of field programmable gate arrays (FPGAs) to accelerate relational joins. Relational join is one of the most CPU-intensive, yet commonly used, database operations. Hashing can be used to reduce the time complexity from quadratic (naive) to linear time. However, doing so can introduce false positives to the results which must be resolved. We present a hash-join engine on FPGA that performs hashing, conflict resolution, and joining on a PCIe-attached system, achieving greater than 11x speedup over software.

引用

页码：17 / 20

页数：4

共 14 条

[1]

[Anonymous], ACM SIGMOD

[2]

Johnson R., P INT C VLDB 08

[3] Sort vs. Hash Revisited: Fast Join Implementation on Modern Multi-Core CPUs [J].

Kim, Changkyu ;

Sedlar, Eric ;

Chhugani, Jatin ;

Kaldewey, Tim ;

Nguyen, Anthony D. ;

Di Bias, Andrea ;

Lee, Victor W. ;

Satish, Nadathur ;

Dubey, Pradeep .

PROCEEDINGS OF THE VLDB ENDOWMENT, 2009, 2 (02) :1378-1389

[4]

Krueger J., 2012, P VLDB ENDOWMENT, V5

[5]

Leung K. T., 1999, UC MICRO TECHNICAL R

[6]

Low B., 2011, INT J NEW COMPUTER A

[7]

Mueller R., 2009, VLDB 09

[8]

Mueller R., ACM SIGMOD 2010, P1159

[9]

Mueller R, 2009, ACM SIGMOD/PODS 2009 CONFERENCE, P999

[10]

Sadoghi M., 2012, IEEE INT C DAT ENG I

← 1 2 →