Hacker News Reader

Welcome to benchmarklandia, where you find the top algorithm in an area and write a stripped down version that outcompetes it on a different set of data and get good results.

See: hadoop->spark->spark with infiniband->spark with nvme->timely dataflow on a laptop; csv parsers that don't support scientific notation floats, etc. I'm sure others can give moren interesting examples.

View on HN

Out of (self-interested) curiosity, what are you referring to with the "nvme->timely dataflow on a laptop" reference?

The closest benchmark I could find was the "FASTER State Management for Timely Dataflow" paper from ETH Zurich, but that wasn't run on a laptop.

http://www.frankmcsherry.org/assets/COST.pdf

It's a bit different, right, because that paper presents better algorithms on the same datasets as the work it cites.

Yes it's different. Another part of Benchmarklandia is where researchers make custom algorithms for the same dataset. Benchmarks targeting MNIST is a great example. :)

Reply on news.ycombinator.com