mbhall88/compression_benchmark

Benchmarking fastq compression with generic (mature) compression algorithms

compression_benchmark's Issues

ENH: Plot ratio vs memory for pareto frontier

Great work! One idea for improvement:

A really useful way to visualize compression benchmarks is to plot ratio vs time/memory so one can get a pareto frontier (for each ratio, what's the fastest). Like this:

https://richg42.blogspot.com/2015/11/the-lossless-decompression-pareto.html

This is a good way to abstract away the arbitrary scale of compression levels - level 2 for gzip isn't the same as 2 for xz etc, all we care about is the line in the memory/ratio (or time/ratio) space, then look up what the level is that gets it for the tool that's best.

parallel implementations and clustering of reads

Hello,

there are mature, included in the major Linux distros programs such as pigz, pbizp2 and lbzip2. These compress faster while being on par on compression ratios. The last one (lbzip2) seems to be faster with decompression.

For achieving better compression ratios it does help to cluster reads based on the sequence with clumpify from BBMap. This also tends to speed up a bit with downstream mapping.

Hope it helps

DK

ENH: Benchmark compression/decompression speed/bandwidth

Speed is something where zstd really shines compared to xz, would be cool to have that dimension benchmarked as well!