Hacker News new | ask | show | jobs
Intel CPU Die Topology (jprahman.substack.com)
58 points by stevenzzzzzzz 1123 days ago
3 comments
Cockbrand 1122 days ago
This might be a nitpick, but it seems like already the intro sentence isn't factually accurate:

> Over the past 10-15 years, per-core throughput slowed down

I assume that the author meant that the acceleration of per-core throughput over time has slowed down, not the throughput itself.

link
dan-robertson 1122 days ago
Why is the rtt measured here on canon lake so asymmetric? I would have expected symmetry along the leading diagonal.
link
yvdriess 1122 days ago
Speculation, but could it be because of a ring topology?
link
jeffbee 1122 days ago
And why is core 0 magical on that machine?
link
jcranmer 1122 days ago
IIRC, core 0 is the one that tends to service processor interrupts, which may mean that the data is unusually noisy in a way that tends to underestimate the RTT latency time.
link
jeffbee 1122 days ago
There’s a similar weird artifact at index 64.
link
innocenat 1122 days ago
Core 64 is the hyper-threaded pair of core 0.
link
jeffbee 1122 days ago
Not on that machine. That would only be true on a 128-thread system, but this one has 96 threads.
link
_a_a_a_ 1122 days ago
> Over the past 10-15 years, per-core throughput slowed down

throughput of what?

link