Hacker News new | past | comments | ask | show | jobs | submit login

Why would the evaluation charts look the way they do?

https://github.com/fastos/fastsocket#online-evaluation

The "before and after" CPU series have nearly the same exact fit. If the data was from separate 24 hour periods, wouldn't you expect the graphs to look different? I recognize that with a large service, you'd get repetitive load patterns, but the similarity here look a little extreme.




I'm guessing here, but it could be two machines, load-balanced. One running with the patch and the other without it.


I agree; it's probably just a language barrier instead of something nefarious.

    In the figure below, it is the CPU utilization
    of a 8-core servers within 24 hours. 
Author probably meant to write "two 8-core servers".


I find the first graph peculiar on its own. Supposedly, each line is the load on one of 8 cores on the same machine. Why would some cores experience heavier load than others, very consistently, over the course of a day? I've never seen a workload exhibit that kind of long-term, core-level affinity on Linux.


This sounds like RSS or flow director gone horribly wrong.


Well, the obvious reason for such a graph is that the network load balancing between several waiting worker processes isn't symmetrical.


Even if that was the case, there isn't normally a stable mapping between processes and physical cores. There would have to be something within the kernel itself that gives higher priority to some cores than others.

Not saying that's impossible, but I've worked on machines with more than 8 cores and never seen it happen.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: