13

AlphaDev uncovered new sorting algorithms that led to improvements in the LLVM libc++ sorting library that were up to 70% faster for shorter sequences and about 1.7% faster for sequences exceeding 250,000 elements.

you are viewing a single comment's thread
view the rest of the comments
[-] Sibbo@sopuli.xyz 2 points 1 year ago

That is 70% faster on sequences of five elements. Using this for handling the end of the recursion in LLVMs current std::sort implementation results in a 1.7% speedup on sequences exceeding 250k elements.

[-] greysemanticist@lemmy.one 2 points 1 year ago

I wonder if their paper has a plot of the speedups against number of elements. Did they just stop measuring at 250k? What was the shape of the curve?

[-] Sibbo@sopuli.xyz 1 points 1 year ago

I didn't bother reading, not so interested in AI stuff. But since they highlight these results, it likely does not get better for higher numbers.

[-] greysemanticist@lemmy.one 1 points 1 year ago

I'll bet it just ends up being the limitations of memory bandwidth to stuff things into registers for the optimized algorithm. Or, something like Mojo's autotuning finds the best way to partition the work for the hardware.

this post was submitted on 08 Jun 2023
13 points (100.0% liked)

Programming

13269 readers
3 users here now

All things programming and coding related. Subcommunity of Technology.


This community's icon was made by Aaron Schneider, under the CC-BY-NC-SA 4.0 license.

founded 1 year ago
MODERATORS