| 21 Aug 2025 |
 K900 | And we can't fan out early bootstrap | 15:44:32 |
| K900 | I'd expect like, 2-3 days overhead per staging cycle from this, which is BAD | 15:44:47 |
| K900 | If my estimates of single machine RISC-V perf are anywhere close to reality | 15:45:06 |
 Alyssa Ross | Couldn't we just not block staging on RISC-V? | 15:45:28 |
| K900 | We could run a separate jobset, but then the same thing could be done outside of official ifnra | 15:45:57 |
| K900 | * We could run a separate jobset, but then the same thing could be done outside of official infra | 15:45:58 |
| K900 | (and probably should be done outside official infra, at least initially) | 15:46:06 |
 Vladimír Čunát | Trust is needed. People who have physical access to the HW can more-or-less put anything into the cache. | 15:53:05 |
| Vladimír Čunát | (though not trivially at least) | 15:53:34 |
 Tristan Ross | If the infra team doesn't have physical access but the hardware team does, is that acceptable for trust? | 15:53:55 |
| K900 | Would not be great | 15:54:38 |
| K900 | But also, again, I want to first see proof that there's enough hardware | 15:54:38 |
| K900 | And that there's enough demand to justify the Hydra costs | 15:54:43 |
| Tristan Ross | What sort of costs does it make on Hydra? Is it just eval and storage? | 15:55:56 |
| K900 | Queue runner is slow | 15:56:35 |
 emily | cache for the entire jobset on a new platform is substantial (since we pay the whole cost every staging cycle) | 15:56:40 |
