Skip to content

Latest commit

 

History

History
19 lines (12 loc) · 1.18 KB

README.md

File metadata and controls

19 lines (12 loc) · 1.18 KB
Raw

Data speed tests

This repository contains some very basic/rough/scruffy scripts for doing back-of-the-envelope benchmarking for various data operations we use in the HiP-CT project.

Contributions welcome if you'd like to see something else on one of the graphs, or a similar graph. Please open an issue with the request instead of a PR, so we can keep the results consistent.

Results

All rates are given in data/time, so larger is better on the y-axis.

Configuration:

  • JPEG2000 files are written with a compression ratio of 10.

Reading/writing different data formats

Figure showing a benchmark of reading/writing different file formats. x-axis is number of pixels and y-axis is speed. TIFF is the fastest for both reading/writing, JPEG2000 is orders of magnitude slower. Used glymur for JPEG2000s and tifffile for TIFFs.

Reading/writing JPEG2000 files

Figure showing a benchmark of reading/writing JPEG2000 files. x-axis is number of pixels and y-axis is speed. glymur is the fastest for both reading/writing, SimpleITK is second for reading (it cannot write), and opencv is slowest for both reading and writing.