The gmodel keyword supports the following formats:gmodel=model_nameRequests GPUs with the specified brand andmodel name (for example, TeslaK80).gmodel=short_model_nameRequests GPUs with a specific brand name (forexample, Tesla, Quadro, NVS) or model type name(for example, K80, P100).gmodel=model_name-mem_sizeRequests GPUs with the specified brand name andtotal GPU memory size. The GPU memory sizeconsists of the number and its unit, whichincludes M, G, T, MB,GB, and TB (for example,12G).
What are the valid model_name and short_model_name, e.g. for our A100 PCIE 40GB GPUs? In bhosts -gpu they appear as "UnknownNVIDIAA100_PCIE_40GB", and the only two strings LSF accepts for gmodel are the long name as in the bhosts output, and "NVIDIAA100_PCIE_40GB". I would have expected something like "NVIDIAA100" or "A100" to work, too, but those jobs will never start, with the pending reason "Not enough GPUs with the required GPU model on the hosts". Is there a way, to make LSF show the "valid" model names?