Skip to content

Latest commit

 

History

History
25 lines (15 loc) · 751 Bytes

benchmarks.md

File metadata and controls

25 lines (15 loc) · 751 Bytes
Raw

First iteration

PR: [https://github.com/Dipenbhatt03/py-db/pull/4/files]
Adding 1 lakh rows take 85 seconds i.e 1171 row/sec
searching worst case takes 2.39 seconds due to linearity

Second iteration

PR: [https://github.com/Dipenbhatt03/py-db/pull/2/files]

img.png Implmenting binary tree for storage and inserting fairly random data, searching speed dramatically increases to 0.002 seconds

Third Iteration

PR: https://github.com/Dipenbhatt03/py-db/pull/5

Due to the implementation of standard binary tree, insertion on skewed data had become extremely slow and same goes for searching during worst case.

After AVL implementation and some Row class optimisation the performance jumped to 28 seconds.