John Ousterhout
John Ousterhout
I’d be happy to send you my kernel patches, but I am off the grid for the next couple of weeks without my laptop. I’ll try to do this when...
I'm back on the grid, and I'm attaching to this message a git patch containing all of my instrumentation changes to the Linux kernel. These are for Linux 6.1.38; if...
Weird... I see the attachment in my copy of the email, but I don't see it on GitHub. I'm trying again, this time sending the response directly from GitHub, rather...
Hi Bjorn, The changes I made to the Linux kernel are entirely for instrumentation; there are no changes required for Homa to operate correctly and I haven't made any performance...
First, the compilation errors: * It looks like the file kernel/timetrace.c isn't including the right header file to declare socket structures such as sockaddr_in6. You'll need to figure out which...
I suspect you are running into NIC issues. Check out the paragraph starting with "NIC support for TSO" in the README.md file. What model of NICs are you using? -John-...
Thanks for the additional information. At this point I think we need to consider one or two very specific data points and drill down to figure out exactly what's happening....
Thanks for the additional information; I think I'm starting to understand the experiment. I'm assuming that the metric of interest for you is "Overall for homa_w2 experiment" line in each...
I have figured what is causing the performance difference. The settings for the benchmark specify 3 client threads to generate the request stream. This turns out to be just barely...