I find that the most interesting systems content here sometimes gets overlooked. In this case, I really enjoyed the combination of new decompression algo + pra

I find that the most interesting systems content here sometimes gets overlooked. In this case, I really enjoyed the combination of new decompression algo + practical example of how to use the new algo to [further] exceed hardware memory bandwidth limits. Seeing a successful CTO mine HN for new algos is cool too.