Intel CPU Die Topology
jprahman.substack.com
Intel CPU Die Topology
1–10 of 11 posts
Re: Intel CPU Die Topology
#2Why is the rtt measured here on canon lake so asymmetric? I would have expected symmetry along the leading diagonal.
Re: Intel CPU Die Topology
#3[dead]
Re: Intel CPU Die Topology
#4Why is the rtt measured here on canon lake so asymmetric? I would have expected symmetry along the leading diagonal.
And why is core 0 magical on that machine?
Re: Intel CPU Die Topology
#5Why is the rtt measured here on canon lake so asymmetric? I would have expected symmetry along the leading diagonal.
And why is core 0 magical on that machine?
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.
Re: Intel CPU Die Topology
#6Earlier quoted context omitted.
And why is core 0 magical on that machine?
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.
There’s a similar weird artifact at index 64.
Re: Intel CPU Die Topology
#7Why is the rtt measured here on canon lake so asymmetric? I would have expected symmetry along the leading diagonal.
Speculation, but could it be because of a ring topology?
Re: Intel CPU Die Topology
#8This 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.
Re: Intel CPU Die Topology
#9Earlier quoted context omitted.
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.
There’s a similar weird artifact at index 64.
Core 64 is the hyper-threaded pair of core 0.
Re: Intel CPU Die Topology
#10> Over the past 10-15 years, per-core throughput slowed down
throughput of what?