Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/srijanshetty/bplustree
A C++ implementation of B+ Tree
https://github.com/srijanshetty/bplustree
btree cpp databases storage
Last synced: 2 months ago
JSON representation
A C++ implementation of B+ Tree
- Host: GitHub
- URL: https://github.com/srijanshetty/bplustree
- Owner: srijanshetty
- License: other
- Created: 2015-02-08T14:26:59.000Z (almost 10 years ago)
- Default Branch: master
- Last Pushed: 2015-03-26T08:22:44.000Z (over 9 years ago)
- Last Synced: 2024-04-14T07:43:41.771Z (8 months ago)
- Topics: btree, cpp, databases, storage
- Language: C++
- Homepage:
- Size: 24.7 MB
- Stars: 13
- Watchers: 3
- Forks: 6
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# B+ Tree
## Results
- The computed time is in microseconds.
### Page Size : 2048kB
QUERY | MAX | MIN | AVG | STD
0 | 4177465 | 95 | 208859 | 20874.2
1 | 389 | 57 | 817.42 | 75.042
2 | 534 | 58 | 1047.53 | 95.453
3 | 1025 | 71 | 1856.51 | 170.451
4 | 1078 | 60 | 1898.71 | 173.971
## Observations
### Insertion vs Other Queries
- The maximum time taken by an insertion is huge compared to the maximum times
of other operations. An insertion can start a chain of insertions when a child
delegates an insertion to its parent, there by accounting for the large maximum.
### Huge Standard Deviation in Insertions
- The huge standard deviation in insertions is accounted by the fact that not
all insertions will cascade into parent insertions.
### Point Queries
- Point Queries take the least amount of time because they only need a single
pass through the tree.
### Range Queries vs Other Queries
- Due to the query bias - range queries had a very small radius, the times reported
for range queries seem to be very small. This shouldn't be the case as internally
rangeQuery calls windowQuery.
### Effectiveness of B+Tree
- One of the most important things to note is the time taken by all these queries is
very small for a disk based structure which shows the effectiveness of B+Tree and the
rationale for using it in Databases.
## INSTALL
- To build a new tree you can run:
```shell
$ make build
```
- To build the tree which was computed in the first run:
```shell
$ make restore
```
- To time the program the configuration is as follows:
```c++
// #define OUTPUT
#define TIME
```
- To just get the output, the configuration is as follows:
```c++
#define OUTPUT
// #define TIME
```