The E10k reminds me of Bryan Cantrill's story about the motivation for dtrace. Sun engineers were working day and night, trying to debug what seemed to be a Solaris kernel networking bug, on a benchmarking cluster of E10k machines. I won't spoil the end, but it's great:
alt Hacker News
The E10k reminds me of Bryan Cantrill's story about the motivation for dtrace. Sun engineers were working day and night, trying to debug what seemed to be a Solaris kernel networking bug, on a benchmarking cluster of E10k machines. I won't spoil the end, but it's great:
https://www.youtube.com/watch?v=wTVfAMRj-7E&t=2640s