datafusion-contrib / datafusion-contrib/liquid-cache
When to train compressors?
Open
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 452
- Forks
- 51
- Avg merge
- 3h 10m
- Merged PRs (30d)
- 12
Description
As @nikhilanayak pointed out we can train compressor for every row group, or every file, or every batch. Finer grained compressors get better compression ratio, but at the cost of extra training. We'd like to quantify this trade-off.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by examining how compressor training is handled at the row-group, file, and batch boundaries mentioned in the issue. Compare the compression results and training cost for each granularity; done means the trade-off between compression ratio and training effort is quantified.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- data-engineering, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 30/100