News for hackers. Stuff that matters.

A comment on Is sequential IO dead in the era of the NVMe drive?. Parent comment

Re:Is sequential IO dead in the era of the NVMe drive? by loeg on Tuesday May 09, 2023 @05:55PM (#35880881)
> So sequential I/O is perhaps half-dead; there's no need to optimize layouts for sequential reads any more.

Yes and no. Large contiguous IOs are still faster to read than a bunch of random small sectors. You generally want your frequently read files/blobs to be split into as few operations as possible.

https://dl.acm.org/doi/abs/10.1145/3477132.3483593

[ Parent | on HN ]
Re:Is sequential IO dead in the era of the NVMe drive? by kevincox on Wednesday May 10, 2023 @04:52PM (#35893347)
This bears a strong resemblance to RAM and caches where cache lines are relatively large so you want to optimize your algorithms to touch as few "cache lines" as possible. So even in RAM random IO is slower than sequential IO.
[ Parent | on HN ]
Re:Is sequential IO dead in the era of the NVMe drive? by loeg on Wednesday May 10, 2023 @05:05PM (#35893506)
Yeah. Causing additional page accesses can also be expensive due to the TLB impact.
[ Parent | on HN ]