The Lewis Maintenance has been completed. The queue was released yesterday and jobs have been running overnight. The login will open shortly. Important information about running jobs below.The new GPU rack has been installed and one more investor GPU node with 3x V100’s has been added to the cluster. We can now easily add GPU investor nodes (about $34,000 for 3x GPU’s), please contact us for details.There are a number of changes to the partitions and running jobs related to investment below.gpu4:
Users that have not invested in gpu4 can run jobs on the Gpu partition (all GPU’s in the system) for up to 2 hours and by request jobs up to 2 days on the gpu3 partition (1080ti, K40, K20m and 1080ti’s). The gpu4 partition for jobs over 2 hours is limited to investors only.GPU investors must now submit jobs to the gpu4 partition (–partition gpu4) with a gpu4 QOS (–qos gpu4) and a GPU investment account to charge (–account engineering-gpu) other than general.
Changes were made to the BioCompute partition which require a user to be a member of the biocompute group and also specify a slurm account with -A <account name> other than general in order for jobs to be run on the partition. If you question whether or not you are a member of the biocompute group you can check using
id $USER | grep biocompute. If that returns a result then you are a member. If not, see  http://docs.rnet.missouri.edu/policies/#biocompute-group-policy
— Nov. 12, 2019