ClassAccounts: Difference between revisions
| Line 99: | Line 99: | ||
<pre> | <pre> | ||
$ show_nodes -p class | $ show_nodes -p class | ||
NODELIST CPUS MEMORY AVAIL_FEATURES GRES STATE | NODELIST CPUS MEMORY AVAIL_FEATURES GRES STATE | ||
tron06 16 126214 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle | tron06 16 126214 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle | ||
Revision as of 16:37, 16 December 2024
Overview
UMIACS Class Accounts support classes for all of UMIACS/CSD via the Nexus cluster. Faculty may request that a class be supported by following the instructions here.
Getting an account
Your TA or instructor will request an account for you. Once this is done, you will be notified by email that you have an account to redeem. If you have not received an email, please contact your TA or instructor. You must redeem the account within 7 days or else the redemption token will expire. If your redemption token does expire, please contact your TA or instructor to have it renewed.
Once you do redeem your account, you will need to wait until you get a confirmation email that your account has been installed. This is typically done once a day on days that the University is open for business.
Any questions or issues with your account, storage, or cluster use must first be made through your TA or instructor.
Registering for Duo
UMIACS requires that all Class accounts register for MFA (multi-factor authentication) under our Duo instance (note that this is different than UMD's general Duo instance). You will not be able to log onto the class submission nodes until you register.
If you see the following error in your SSH client, you have not yet enrolled/registered in Duo.
Access is not allowed because you are not enrolled in Duo. Please contact your organization's IT help desk.
In order to register, visit our directory app and log in with your Class username and password. You will then receive a prompt to enroll in Duo. For assistance in enrollment, please visit our Duo help page.
Once notified that your account has been installed and you have registered in our Duo instance, you can SSH to nexusclass.umiacs.umd.edu with your assigned username and your chosen password to log in to a submission node.
If you store something in a local filesystem directory (/tmp, /scratch0) on one of the two submission nodes, you will need to connect to that same submission node to access it later. The actual submission nodes are:
nexusclass00.umiacs.umd.edunexusclass01.umiacs.umd.edu
Cleaning up your account before the end of the semester
Class accounts for a given semester are liable to be archived and deleted after that semester's completion as early as the following:
- Winter semesters: February 1st of same year
- Spring semesters: June 1st of same year
- Summer semesters: September 1st of same year
- Fall semesters: January 1st of next year
It is your responsibility to ensure you have backed up anything you want to keep from your class account's personal or group storage (below sections) prior to the relevant date.
Personal Storage
Your home directory has a quota of 30GB and is located at:
/fs/classhomes/<semester><year>/<coursecode>/<username>
where <semester> is either "spring", "summer", "fall", or "winter", <year> is the current year e.g., "2021", <coursecode> is the class' course code as listed in UMD's Schedule of Classes in all lowercase e.g., "cmsc999z", and <username> is the username mentioned in the email you received to redeem the account e.g., "c999z000".
You can request up to another 100GB of personal storage if you would like by having your TA or instructor contact staff. This storage will be located at
/fs/class-projects/<semester><year>/<coursecode>/<username>
Group Storage
You can also request group storage by having your TA or instructor contact staff to specify the usernames of the accounts that should be in the group. Only other class accounts in the same class can be added to the group. The quota will be 100GB multiplied by the number of accounts in the group and will be located at
/fs/class-projects/<semester><year>/<coursecode>/<groupname>
where <groupname> is composed of:
- the abbreviated course code as used in the username e.g., "c999z"
- the character "g"
- the number of the group (starting at 0 for the first group for the class requested to us) prepended with 0s to make the total group name 8 characters long
e.g., "c999zg00".
Cluster Usage
You may not run computational jobs on any submission node. You must schedule your jobs with the SLURM workload manager. You can also find out more with the public documentation for the SLURM Workload Manager.
Class accounts only have access to the following submission parameters in SLURM:
--partition-class--account-class--qos-default,medium, andhigh
You must specify at least the partition parameter manually in any submission command you run. If you do not specify any QoS parameter, you will receive the QoS default.
You can view the resource limits for each QoS by running the command show_qos. The value in the MaxWall column is the maximum runtime you can run a single job for each QoS, and the values in the MaxTRES column are the maximum amount of CPU cores/GPUs/memory you can request for a single job using each QoS.
Please note that you will be restricted to 32 total CPU cores, 4 total GPUs, and 256GB total RAM across all jobs you have running at once. This can be viewed with the command show_partition_qos.
Example
Here is a basic example to schedule a interactive job running bash with a single GPU in the partition class, with the account class, running with the QoS of default and the default CPU/memory allocation/time limit for the partition.
bash-4.4$ hostname nexusclass00.umiacs.umd.edu bash-4.4$ srun --partition=class --account=class --qos=default --gres=gpu:1 --pty bash srun: Job time limit was unset; set to partition default of 60 minutes srun: job 1333337 queued and waiting for resources srun: job 1333337 has been allocated resources bash-4.4$ hostname tron14.umiacs.umd.edu bash-4.4$ nvidia-smi -L GPU 0: NVIDIA RTX A4000 (UUID: GPU-55f2d3b7-9162-8b02-50de-476a012c626c)
Available Nodes
You can list the available nodes and their current state with the show_nodes -p class command. This list of nodes is not completely static as nodes may be pulled out of service to troubleshoot GPUs or other components.
$ show_nodes -p class NODELIST CPUS MEMORY AVAIL_FEATURES GRES STATE tron06 16 126214 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle tron07 16 126214 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle tron08 16 126214 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle tron09 16 126214 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle tron10 16 126217 rhel8,x86_64,Zen,EPYC-7313P,Ampere gpu:rtxa4000:4 idle tron11 16 126217 rhel8,x86_64,Zen,EPYC-7313P,Ampere gpu:rtxa4000:4 idle tron12 16 126218 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle tron13 16 126214 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle tron14 16 126214 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle tron15 16 126214 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle tron16 16 126217 rhel8,x86_64,Zen,EPYC-7313P,Ampere gpu:rtxa4000:4 idle tron17 16 126217 rhel8,x86_64,Zen,EPYC-7313P,Ampere gpu:rtxa4000:4 idle tron18 16 126217 rhel8,x86_64,Zen,EPYC-7313P,Ampere gpu:rtxa4000:4 idle tron19 16 126217 rhel8,x86_64,Zen,EPYC-7313P,Ampere gpu:rtxa4000:4 idle tron20 16 126217 rhel8,x86_64,Zen,EPYC-7313P,Ampere gpu:rtxa4000:4 idle tron21 16 126218 rhel8,x86_64,Zen,EPYC-7302P,Ampere gpu:rtxa4000:4 idle tron22 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron23 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron24 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron25 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron26 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron27 16 126214 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron28 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron29 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron30 16 126214 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron31 16 126214 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron32 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron33 16 126214 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron34 16 126217 rhel8,x86_64,Zen,EPYC-7313P,Ampere gpu:rtxa4000:4 idle tron35 16 126214 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron36 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron37 16 126214 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron38 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron39 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron40 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron41 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron42 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron43 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron44 16 126218 rhel8,x86_64,Zen,EPYC-7302,Ampere gpu:rtxa4000:4 idle tron46 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron47 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron48 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron49 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron50 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron51 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron52 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron53 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron54 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron55 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron56 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron57 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron58 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron59 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron60 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle tron61 48 255232 rhel8,x86_64,Zen,EPYC-7352,Ampere gpu:rtxa5000:8 idle
You can also find more granular information about an individual node with the scontrol show node command.
$ scontrol show node tron06 NodeName=tron06 Arch=x86_64 CoresPerSocket=16 CPUAlloc=0 CPUEfctv=16 CPUTot=16 CPULoad=0.08 AvailableFeatures=rhel8,x86_64,Zen,EPYC-7302P,Ampere ActiveFeatures=rhel8,x86_64,Zen,EPYC-7302P,Ampere Gres=gpu:rtxa4000:4 NodeAddr=tron06 NodeHostName=tron06 Version=23.02.6 OS=Linux 4.18.0-513.11.1.el8_9.x86_64 #1 SMP Thu Dec 7 03:06:13 EST 2023 RealMemory=126214 AllocMem=0 FreeMem=107174 Sockets=1 Boards=1 State=IDLE ThreadsPerCore=1 TmpDisk=0 Weight=340 Owner=N/A MCS_label=N/A Partitions=class,scavenger,tron BootTime=2024-01-29T09:35:12 SlurmdStartTime=2024-02-05T15:14:20 LastBusyTime=2024-02-16T15:59:38 ResumeAfterTime=None CfgTRES=cpu=16,mem=126214M,billing=638,gres/gpu=4,gres/gpu:rtxa4000=4 AllocTRES= CapWatts=n/a CurrentWatts=0 AveWatts=0 ExtSensorsJoules=n/s ExtSensorsWatts=0 ExtSensorsTemp=n/s