Presentation 2002/7/10
Accelerating insertion processing to the CSS-tree
Yoshihisa HONDA, Hideyuki KAWASHIMA, Michita IMAI, Yuichiro ANZAI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In recent years, memory index structures that accelerate search speed utilizing the structure of CPU cache have proposed. Especially, the Full Cache Sensitive Search Tree (CSS-tree) proposed in 1999 is faster than hash depending on data size and require smaller amount of space than B-tree. On the other hand, reconstructing the CSS-tree is so slow that it has been considered that the CSS-tree is only useful under static environments such as OLAP. However, if updating is limited only to additional insertions, the reconstruction cost can be dramatically decreased. Then, in this research, we propose the new techniques that accelerate insertion processing to the C-SS-tree. The technique consists of (1) preliminary acquiring excess index structure space to reduce cost for node expanding and (2) omitting leaf nodes to reduce cost for remapping the leaf nodes and instead, storing key value into internal nodes. To evaluate the proposed technique, we implemented an experimental system with C language and assemblers on SunOS 5.6 and measured the insertion time and the size of memory space for the structure. The result showed that the proposed technique accelerates 7.18 times faster than original technique.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) CSS-tree / database System / memory index structure / cache memory
Paper # DE2002-20
Date of Issue

Conference Information
Committee DE
Conference Date 2002/7/10(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Data Engineering (DE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Accelerating insertion processing to the CSS-tree
Sub Title (in English)
Keyword(1) CSS-tree
Keyword(2) database System
Keyword(3) memory index structure
Keyword(4) cache memory
1st Author's Name Yoshihisa HONDA
1st Author's Affiliation Graduate School of Science and Technology, Keio University()
2nd Author's Name Hideyuki KAWASHIMA
2nd Author's Affiliation Graduate School of Science and Technology, Keio University
3rd Author's Name Michita IMAI
3rd Author's Affiliation Faculty of Science and Technology, keio University
4th Author's Name Yuichiro ANZAI
4th Author's Affiliation Faculty of Science and Technology, keio University
Date 2002/7/10
Paper # DE2002-20
Volume (vol) vol.102
Number (no) 207
Page pp.pp.-
#Pages 6
Date of Issue