Comment on Measuring CPU core-to-core latencyComments−apaolillo3yWe published a paper where we captured the same kind of insights (deep numa hierarchies including cache levels, numa nodes, packages) and used them to tailor spinlocks to the underlying machine: https://dl.acm.org/doi/10.1145/3477132.3483557−nviennotOP3yVery nice
Comments
We published a paper where we captured the same kind of insights (deep numa hierarchies including cache levels, numa nodes, packages) and used them to tailor spinlocks to the underlying machine: https://dl.acm.org/doi/10.1145/3477132.3483557
Very nice