THE DECK · OPS LAYER FOR GPU OPERATORS
If you supply GPUs to rental networks, you know the failure modes: a worker's memory climbs until the kernel kills it mid-render, destroying the paid job, and the container exits clean so nothing even tells you. A lane drops after an image update and earns nothing for a day. Dashboards say the process is healthy while the payout says otherwise. The deck is the operations layer for exactly this - built by an operator, for operators, running against a real production fleet every night.
Workers restart in an idle window, before the memory cap - measured on anon-rss, not the cache-polluted number container stats show you. A recycle costs one idle restart; the kill costs the restart plus the paid job it destroyed.
Checks verify at the payout layer, never the process layer. A green container that stopped earning is an incident; the deck treats it as one before you notice.
Lane down, model set wrong after a reboot, image update stalled - the recovery runs first and the report arrives after, with what failed and what fixed it.
Memory limits, headroom, routing - rolled out as controlled comparisons with a pre-registered read-out, reverted automatically when the control wins.
ISSUE lane-726 anonymous memory at 41% of its cgroup cap and climbing, 24 minutes into a paid shift. The unmanaged path ends in a cgroup kill mid-job: one failed render, one worker restart, minutes of paid occupancy gone.
INTERVENTION recycled at 01:23:57 inside a 51-second idle window between jobs — chosen, not lucky: the deck had been watching for that window since 01:23:04.
AVOIDED the kill path above. No earning figure is claimed for this case — the deck records the avoidance, not a hypothetical recovery number.
EVIDENCE the run log, verbatim, lane ids relabelled:
01:23:04 memory: lane-11=58% lane-726=41% lane-9=24% lane-725=22% 01:23:04 lane-726: 41% of cap (19.6 GiB anon, up 24 min) - waiting for an idle window 01:23:57 lane-726: RECYCLED before the cgroup could kill it (51s, between jobs) 01:26:05 nothing at threshold; no recycle needed
Real log lines from last night's production run, lane ids relabelled. The job that render was carrying finished and paid.
The deck runs nightly on the fleet behind the GPUDeck record. Productizing it for operators on Vast, Clore, Sogni and similar networks is next: self-serve, your machines, your keys, no agent phoning home. If that is your situation, say hello - cohort invitations go out in order.
You're on the list - cohort invitations go out in order. A person replies from deck@gpudeck.com.
That email didn't validate - try again, or write to deck@gpudeck.com directly.
Prefer mail? deck@gpudeck.com with subject “GPUDeck deck waitlist” works too.
Honesty, stated plainly: there is no product to download yet. The tooling exists and runs in production for one fleet; the waitlist decides whether it becomes yours too. No spam, no list resale - replies come from a person.