Linux IO_uring Can Now Achieve Up To ~3.8M IOPS Per-Core

It was just last month when ~3.5M IOPS per-core was impressive with the code for Linux 5.15 to further push Linux's I/O limits. Now for code likely to be included in Linux 5.16 it's currently at 3.8M IOPS with a single tread.

With this patch series reworking and further optimizing the submission and completion paths, the I/O throughput is upped even more. With block maintainer and IO_uring lead developer Jens Axboe's Intel Optane based rig, he is enjoying around a 3% throughput improvement.

Axboe tweeted that with these latest patches he is now achieving 3.8M IOPS on a single core.

Pavel continues his relentless quest to squeeze every bit of efficiency out of the stack. This one nets me about a 3% improvement in single thread performance, 3.7M -> 3.8M IOPS.https://t.co/bzWQGZ7GLG

— Jens Axboe (@axboe) September 25, 2021

Will 4M IOPS on a single thread be achievable this year? We'll see... Lots of great work happening in IO_uring land and throughout Linux's storage landscape.