Cache Benchmarks

Re:Cache Benchmarks by phoronixrly on Tuesday July 15, 2025 @06:26AM (#44569724)

> c8g.8xlarge (32 core non-NUMA ARM64)

Requests are scheduled on half of these. Despite that, a plateau is hit after 8 threads? Is this a 16-core 32-thread type of a setup?

Also, consider redoing this in linear scale.

Edit: Oddly enough, no? 1 thread per core as per https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/cpu-opti...

[ Parent | on HN ]

Re:Cache Benchmarks by swiftcoder on Tuesday July 15, 2025 @07:34AM (#44570079)

> a plateau is hit after 8 threads?

Most of the graphs plateau around 6 threads, for pretty much all the caches under test. I wonder if there is some interesting architectural issue with cache-sharing on this particular platform?

[ Parent | on HN ]

Re:Cache Benchmarks by namibj on Tuesday July 15, 2025 @09:18AM (#44570847)

I guess memory controller behavior; especially if it's not set up for parallelism IOPS but for single-threaded throughout (channel interleaving).

[ Parent | on HN ]

Re:Cache Benchmarks by everfrustrated on Tuesday July 15, 2025 @08:18AM (#44570366)

g = graviton which doesn't support SMT so 1 vcpu is 1 full core

[ Parent | on HN ]

Re:Cache Benchmarks by tidwall on Thursday July 17, 2025 @12:34PM (#44595152)

Thanks for the feedback. The plateauing is due to a taskset misconfiguration. I fixed it and reran the benches. They look much better now.

[ Parent | on HN ]