Orasort: 5x faster column-sorting with an expired patent from Oracle

dafelst

It's kind of insane that such an obvious optimization can be patented, I have to imagine that it has been invented independently dozens if not hundreds of times.

beastman82

patent: https://patents.google.com/patent/US7680791B2

charcircuit

>A CPU register is naturally 8 bytes in size What does naturally even mean here. How is a 64 byte register's (zmm0) size any less natural?

orlp

First, this article is mostly (AI?) regurgitation. This is much better: https://smalldatum.blogspot.com/2026/01/common-prefix-skippi... . Second, I have independently invented this (quicksort on string prefixes) at my time at CWI, although I didn't end up publishing it, because... Third, this was already published in the original 1961 Quicksort paper by Hoare: https://www.cs.ox.ac.uk/files/6226/H2006%20-%20Historic%20Qu... . Near the end, the section on "Multi-word keys" describes a quicksort that partitions on just the first word, and only accesses the next word for the equality partition. And funnily enough this paper credits P. Shackleton for this, thus this idea was thought of even before the Quicksort paper came out. So as is usual for software patents, this patent never should have been awarded.

hermitcrab

A vague article. With one sentence per line. Most annoying.

Validark

Thank you. For wasting my time. The only thing someone could learn from this is that CPU registers can be 8 bytes.

galkk

Looks like an AI rewrite of something better.

ranger_danger

> A CPU register is naturally 8 bytes in size, and if Oracle extracts 8 bytes from two strings for comparison, then the comparison requires fewer registers and fewer CPU cycles. Isn't this just a typical SIMD optimization that tons of projects use?