Serializing orthogonal range (k-d) trees for fast on-disk queries
Switch branches/tags
Nothing to show
Clone or download
Pull request Compare This branch is 10 commits ahead, 2 commits behind ktsiegel:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
ort
serializer
.DS_Store
.gitignore
PKG-INFO
README.md
fig1.eps
fig2.eps
fig3.eps
hodor.pdf
hodor.tex
requirements.txt

README.md

HODOR: HODOR On-Disk Orthogonal Range-trees

The orthogonal range tree is a data structure for efficient range queries on d dimensions. HODOR is an on-disk implementation introducing the idea of 'back seek' complexity, or complexity in the number of reads of locations falling before the current disk position. By serializing the tree in a particular way, HODOR achieves a constant number of back seeks per range query. Check out hodor.pdf for more details.