Damian Edwards

I think the math is off, as networking bandwidth is measured as base 1000 for Kbs -> Mbps -> Gbps, unlike storage which is 1024. So the theoretical max for...

I believe the request headers for plaintext are currently the following: https://github.com/TechEmpower/FrameworkBenchmarks/blob/e99d22fdd6cf10354e6dd6d4422c64dfd5f28cc2/toolset/benchmark/test_types/plaintext_type.py#L72 That would make the full request the following: ``` GET /plaintext HTTP/1.1 Host: tfb-server Accept: text/plain,text/html;q=0.9,application/xhtml+xml;q=0.9,application/xml;q=0.8,*/*;q=0.7 Connection: keep-alive...

As an aside, it seems possible we're also starting to hit limits in the JSON test too, although in that case it seems aligned with the packet switching limit.

Sounds good. 100G fiber here we come 😁

Yeah the math was just meant to be HTTP only theoretical max, to see if we're in the ballpark, and it's clear we are. Would be interesting to include overheads...

The point raised regarding testing TLS handshake performance (aka for new connections) vs. encoding/encryption costs is an interesting one, and makes me wonder whether we should configure the Fortunes variant...

This would be great info, along with BIOS version, OS, etc. given our recent efforts to try and ensure our environments match (and apparently failing). /cc @sebastienros

Perhaps we could keep this info in an MD file in this repo and just link to it from the environment section of the public page, in an effort to...

@msmith-techempower for cloud I'd suggest we simply state what VM SKUs are being used.

Seems the URL the FrontEnd uses to speak to the BackEnd is misconfigured. Ensure that the BackEnd is listening on the URL you have configured for the FrontEnd (in the...