18How many spare GPUs, nodes, cables and transceivers do you hold for a 2,048-GPU fleet?▼mediumNewCoreWeaveMetaLambda Labs5 replies○ sign inThe spares number falls out of the failure rate times the repair turnaround, and the cheap items are the ones people forget. The arithmetic for GPUs, the very different arithmetic for transceivers, and the rule that decides whether to hold a part at all.Open full answer →
37Design the system that tells you every firmware version on every device in the fleet.▼mediumNewMetaCoreWeaveMicrosoft4 replies◆ premiumFirmware skew produces intermittent faults that surface layers away from their cause, so an inventory is a debugging tool rather than a compliance exercise. What to collect, how drift appears without anyone deploying anything, and the diff that turns a week-long investigation into a query.Open full answer →