What an occupied GPU hour finishes

Replication. Every cell on this page comes from a single operator and a single day. 0 of the 11 rows below rest on one physical card on at least one side, marked ●. Across the whole table 11 distinct cards contributed. Attempt counts are not replication: a card that ran 600 attempts in a day is still one card, and confidence that treats those attempts as independent would be wrong by roughly the square root of that count.

The headline does not rest on a single card. Dropping every row that has only one physical card on a side leaves 11 rows and the same range, 1.70x to 3.60x. The thin rows sit inside the range, not at its edges, so they are not setting it.

A deployment here is a GPU together with the host, virtualisation layer, software stack and dispatch it ran under — not the card in isolation. These ratios do not isolate a hardware effect, which is why the confounded rows are separated above rather than averaged in.

Goodput is measured on production hosts, and the host is inside the measurement. The class anchoring the widest ratios above runs under a virtualisation layer that reserves part of its memory and cannot be configured away, so treat those ratios as a result that includes a known hosting disadvantage rather than as a property of the hardware alone. The equal-memory table, where both classes sit on comparable hosts, is the cleaner contrast.

Stated per workload and never combined. A single figure across workloads would move with the mix of work each class happens to be sent, which is a fact about dispatch rather than about the hardware — the same substitution this record exists to refuse. Yield is recorded as production loss under the observed network regime, not as a property of any class.

Derived from the goodput table above — no additional observations. This is the same set of measured pairs, reinterpreted as a price threshold rather than an output ratio, so it is a second reading of one body of evidence and not a second, independent result.

Read every row as a threshold, not a price. A multiple of 3.60x on a row says the larger-memory GPU is the cheaper way to finish that workload at any hourly rate below 3.60 times the smaller one; at exactly that multiple the two cost the same. The daggered rows are not usable this way, because part of their headroom is a hosting penalty specific to this fleet.

Read it as a threshold, not a price. A ratio of 4.00x says the larger class is the cheaper way to finish this workload at any hourly multiple below four times; above it, the smaller class wins. The comparison holds where both classes see the same share of billed time spent occupied — where utilisation differs, scale each side by its own occupied-to-billed fraction before comparing.

Pairs are fixed by largest memory, then lowest permanent class index, both chosen before any measurement was read. Selecting the fastest and slowest observed classes instead would pick each comparison on its own outcome. These 11 rows are drawn from 1 distinct GPU-model pairings, so they show how a small number of comparisons behave across many workloads rather than that many independent comparisons; the GPUs in each row are named so this is checkable.