| 11 Oct 2024 |
 Mindavi | To be clear, the main bottleneck is compression for uploading the build results to S3? Or am I misunderstanding? | 04:51:08 |
 Tristan Ross | In reply to @emilazy:matrix.org
if the Foundation will pay for a month then it seems worth it to just do it? Tbf, I don't think the cost is much of a concern, Tom brought up the cost being not too much more and I've been browsing Hetzner's site and the price difference is like 100 euros. | 04:51:56 |
| Tristan Ross | In reply to @rick:matrix.ciphernetics.nl
To be clear, the main bottleneck is compression for uploading the build results to S3? Or am I misunderstanding? From what I've read, it seems compression. | 04:52:07 |
| Tristan Ross | I think S3 uploads would be more likely to saturate a network rather than cause CPU throttling | 04:52:45 |
 emily | I think both compression and uploads are bottlenecks? | 04:56:29 |
| emily | not sure though. | 04:56:45 |
| emily | "waiting for outputs", is what I always here. | 04:56:50 |
| Tristan Ross | Oh | 04:57:04 |
| emily | * "waiting for outputs", is what I always hear. | 04:57:06 |
| Mindavi | Isn't that compression on the builders? | 05:11:06 |
| Tristan Ross | In reply to @rick:matrix.ciphernetics.nl
Isn't that compression on the builders? No | 05:11:24 |
| Tristan Ross | https://github.com/NixOS/hydra/issues/553 | 05:11:35 |
| Mindavi | Ah, like that | 05:14:43 |
| Mindavi | Ok now it makes sense again :) | 05:14:55 |
 vcunat | I'm fairly certainly that the compression of results on the central machine is what's limiting the throughput almost all the time, as the farm is currently. | 06:11:34 |
| emily | an EPYC with better thermals seems like it'd be a good investment, then? | 06:12:06 |
| vcunat | We have some bigger aarch64 in the farm already, so alternatively we could try that "for free". | 06:13:08 |
| emily | wouldn't that run into the atomics and platform purity problems wrt the evaluator? | 06:13:59 |
