TALC Cluster Guide

From RCSWiki
Revision as of 18:30, 24 February 2021 by Lleung (talk | contribs)
Jump to navigation Jump to search

Security Icon.png

Cybersecurity awareness at the U of C

Please note that there are typically about 950 phishing attempts targeting University of Calgary accounts each month. This is just a reminder to be careful about computer security issues, both at home and at the University. Please visit https://it.ucalgary.ca/it-security for more information, tips on secure computing, and how to report suspected security problems.

This guide gives an overview of the Teaching and Learning Cluster (TALC) at the University of Calgary and is intended to be read by new account holders getting started on TALC. This guide covers topics as the hardware and performance characteristics, available software, usage policies and how to log in and run jobs.

Introduction

TALC is a cluster of computers created by Research Computing Services in response to requests for a central computing resource to support academic courses and workshops offered at the University of Calgary. It is a complement to the Advanced Research Computing (ARC) cluster that is used for research, rather than educational purposes. The software environment in the TALC and ARC clusters very similar and workflows between the two clusters are identical. What students learn about using TALC will have direct applicability to using ARC should they go on to use ARC for research work.

If you are the instructor for a course that could benefit from using TALC, please review this guide, the TALC Terms of Use, then contact us at support@hpc.ucalgary.ca to discuss your requirements. To ensure that the appropriate software is available, student accounts are in place, and appropriate training has been provided for your teaching assistants, it is best to start this discussion several months prior to the start of the course.

If you are a student in a course using TALC, please review this guide for basic instructions in using the cluster. Questions should first be directed to the teaching assistants or instructor for your course.

Obtaining an account

TALC account requests are expected to be submitted by the course instructor rather than from individual students. You must have a University of Calgary IT account in order to use TALC. If you do not have a University of IT account or email address, please register for one at https://itregport.ucalgary.ca/.

Getting Support

Support Icon.png

Need Help or have other TALC Related Questions?

Students, please send TALC-related questions to your course instructor or teaching assistants.
Course instructors and TAs, please report system issues to support@hpc.ucalgary.ca).

Hardware

The TALC cluster is comprised of repurposed research clusters that are a few generations old. As a result, individual processor performance will not be comparable to the latest processors but should be sufficient for educational purposes and course work.

Partition Description Nodes CPU Cores, Model, and Year Installed Memory GPU Network
cpu12 GPU Compute 3 12 cores, 2x Intel(R) Xeon(R) Bronze 3204 CPU @ 1.90GHz (2019) 192 GB 5x NVIDIA Corporation TU104GL [Tesla T4] 40 Gbit/s InfiniBand
cpu24 General Purpose Compute 15 24 cores, 4x Six-Core AMD Opteron(tm) Processor 8431 (2009) 256 GB N/A 40 Gbit/s InfiniBand
bigmem General Purpose Compute 2 32 cores, 4x Intel(R) Xeon(R) CPU E7- 4830  @ 2.13GHz (2015) 1024 GB N/A 40 Gbit/s InfiniBand

Storage

Information Icon.png

No Backup Policy!

You are responsible for your own backups. Since accounts on TALC and related data are removed shortly after the associated course has finished, you should download anything you need to save to your own computer before the end of the course.

TALC is connected to a network disk storage system. This storage is split across the /home and /scratch file systems.

/home: Home file system

Each user has a directory under /home and is the default working directory when logging in to TALC. Each home directory has a per-user quota of 500 GB. This limit is fixed and cannot be increased.

Note on file sharing: Due to security concerns, permissions set using chmod on your home directory to allow other users to read/write to your home directory be automatically reverted by an automated system process unless an explicit exception is made. If you need to share files with other researchers on the ARC cluster, please write to support@hpc.ucalgary.ca to ask for such an exception.

/scratch: Scratch file system for large job-oriented storage

Associated with each job, under the /scratch directory, a subdirectory is created that can be referenced in job scripts as /scratch/${SLURM_JOB_ID}. You can use that directory for temporary files needed during the course of a job. Up to 30 TB of storage may be used, per user (total for all your jobs) in the /scratch file system.

Data in /scratch associated with a given job will be deleted automatically, without exception, five days after the job finishes.

Software

Information Icon.png

Software Package Requests

Course instructors or teaching assistants should write to support@hpc.ucalgary.ca if additional software is required for their course.

All ARC nodes run the latest version of CentOS 7 with the same set of base software packages. For your convenience, we have packaged commonly used software packages and dependencies as modules available under /global/software. If your software package is not available as a module, you may also try Anaconda which allows users to manage and install custom packages in an isolated environment.

For a list of available packages that have been made available, please see ARC Software pages.

Modules

The setup of the environment for using some of the installed software is through the module command.

Software packages bundled as a module will be available under /global/software and can be listed with the module avail command.

$ module avail

To enable Python, load the Python module by running:

$ module load python/anaconda-3.6-5.1.0

To unload the Python module, run:

$ module remove python/anaconda-3.6-5.1.0

To see currently loaded modules, run:

$ module list

Using TALC

Support Icon.png

Usage subject to TALC Terms of Use

Please review the TALC Terms of Use prior to using TALC.

Logging in

To log in to TALC, connect using SSH to talc.ucalgary.ca. Connections to TALC are accepted only from the University of Calgary network (on campus) or through the University of Calgary General VPN (off campus).

See Connecting to RCS HPC Systems for more information.

Working interactively

ARC uses the Linux operating system. The program that responds to your typed commands and allows you to run other programs is called the Linux shell. There are several different shells available, but, by default you will use one called bash. It is useful to have some knowledge of the shell and a variety of other command-line programs that you can use to manipulate files. If you are new to Linux systems, we recommend that you work through one of the many online tutorials that are available, such as the UNIX Tutorial for Beginners (external link) provided by the University of Surrey. The tutorial covers such fundamental topics, among others, as creating, renaming and deleting files and directories, how to produce a listing of your files and how to tell how much disk space you are using. For a more comprehensive introduction to Linux, see The Linux Command Line (external link).

The TALC login node may be used for such tasks as editing files, compiling programs and running short tests while developing programs. We suggest CPU intensive workloads on the login node be restricted to under 15 minutes as per our cluster guidelines. For interactive workloads exceeding 15 minutes, use the salloc command to allocate an interactive session on a compute node.

The default salloc allocation is 1 CPU and 1 GB of memory. Adjust this by specifying -n CPU# and --mem Megabytes. You may request up to 5 hours of CPU time for interactive jobs.

salloc --time 5:00:00 --partition cpu24 

Running non-interactive jobs (batch processing)

Production runs and longer test runs should be submitted as (non-interactive) batch jobs, in which commands to be executed are listed in a script (text file). Batch jobs scripts are submitted using the sbatch command, part of the Slurm job management and scheduling software. #SBATCH directive lines at the beginning of the script are used to specify the resources needed for the job (cores, memory, run time limit and any specialized hardware needed).

Most of the information on the Running Jobs page on the Compute Canada web site is also relevant for submitting and managing batch jobs and reserving processors for interactive work on TALC. One major difference between running jobs on the TALC and Compute Canada clusters is in selecting the type of hardware that should be used for a job. On TALC, you choose the hardware to use primarily by specifying a partition, as described below.

Selecting a partition

TALC currently has the following partitions available for use. The gpu and cpu12 partitions are backed by the same nodes. The cpu12 partition was created to only expose the CPUs on the GPU hardware for general purpose use. Each GPU node has 5 Tesla T4 GPUs installed, but you may only request one per job within the TALC environment.

Partition Description Nodes Cores Memory Memory Request Limit Time Limit GPU Request per Job Network
gpu GPU Compute 3 12 cores 192 GB 190,000 MB 24 hours 1x NVIDIA Corporation TU104GL [Tesla T4] 40 Gbit/s InfiniBand
cpu12 General Purpose Compute 3 12 cores 192 GB 190,000 MB 24 hours None 40 Gbit/s InfiniBand
cpu24 General Purpose Compute 15 24 cores 256 GB 255,000 MB 24 hours None 40 Gbit/s InfiniBand
bigmem General Purpose Compute 2 32 cores 1024 GB 1,000,000 MB 24 hours None 40 Gbit/s InfiniBand

There are some aspects to consider when selecting a partition including:

  • Resource requirements in terms of memory and CPU cores
  • Hardware specific requirements, such as GPU or CPU Instruction Set Extensions
  • Partition resource limits and potential wait time
  • Software support parallel processing using Message Passing Interface (MPI), OpenMP, etc.
    • Eg. MPI for parallel processing can distribute memory across multiple nodes, per-node memory requirements could be lower. Whereas, OpenMP or single process code that is restricted to one node would require a higher memory node.
    • Note: MPI code running on hardware with Omni-Path networking should be compiled with Omni-Path networking support. This is provided by loading the openmpi/2.1.3-opa or openmpi/3.1.2-opa modules prior to compiling.

Since resources that are requested are reserved for your job, please request only as much CPU and memory as your job requires to avoid reducing the cluster efficiency. If you are unsure which partition to use or the specific resource requests that are appropriate for your jobs, please contact us at support@hpc.ucalgary.ca and we would be happy to work with you.

Using a partition

Bigmem and compute-only jobs

To select the cpu24 partition, include the following line in your batch job script:

#SBATCH --partition=cpu24

You may also start an interactive session with salloc:

$ salloc --time 1:00:00 -p cpu24

GPU jobs

In TALC, you are limited to exactly 1 GPU per job. Jobs that request for 0 GPUs or 2 or more GPUs will not be scheduled.

To submit a job using the gpu partition with one GPU request, include the following to your batch job script:

#SBATCH --partition=gpu
#SBATCH --gpus-per-node=1

Like the previous example, you may also request interactive sessions with GPU nodes using salloc. Just specify the gpu partition and the number of GPUs required.

$ salloc --time 1:00:00 -p gpu -n 1 --gpus-per-node 1

You may verify that a GPU was assigned to your job or interactive session by running nvidia-smi. This command will show you the status of the GPU that was assigned to you.

$ nvidia-smi
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 450.80.02    Driver Version: 450.80.02    CUDA Version: 11.0     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  Tesla T4            Off  | 00000000:3B:00.0 Off |                    0 |
| N/A   36C    P0    14W /  70W |      0MiB / 15109MiB |      5%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

Partition limitations

In addition to the hardware limitations of the nodes within the partition, please be aware that there may also be policy limits imposed on your account for each partition. These limits restrict the number of cores, nodes, or GPUs that can be used at any given time. Since the limits are applied on a partition-by-partition basis, using resources in one partition should not affect the available resources you can use in another partition.

These limits can be listed by running:

$ sacctmgr show qos format=Name,MaxWall,MaxTRESPU%20,MaxSubmitJobs
      Name     MaxWall            MaxTRESPU MaxSubmit
---------- ----------- -------------------- ---------
    normal  1-00:00:00          mem=127000M          
     cpu24  1-00:00:00             mem=127G          
    bigmem  1-00:00:00                               
       gpu                       gres/gpu=1

Time limits

Use the --time directive to tell the job scheduler the maximum time that your job might run. For example:

#SBATCH --time=hh:mm:ss

You can use scontrol show partitions or sinfo to see the current maximum time that a job can run.

$ scontrol show partitions
PartitionName=single                                                                 
   AllowGroups=ALL AllowAccounts=ALL AllowQos=ALL                                    
   AllocNodes=ALL Default=NO QoS=single                                              
   DefaultTime=NONE DisableRootJobs=NO ExclusiveUser=NO GraceTime=0 Hidden=NO        
   MaxNodes=UNLIMITED MaxTime=7-00:00:00 MinNodes=1 LLN=NO MaxCPUsPerNode=UNLIMITED  
   Nodes=cn[001-168]                                                                 
   PriorityJobFactor=1 PriorityTier=1 RootOnly=NO ReqResv=NO OverSubscribe=NO        
   OverTimeLimit=NONE PreemptMode=OFF                                                
   State=UP TotalCPUs=1344 TotalNodes=168 SelectTypeParameters=NONE                  
   DefMemPerNode=UNLIMITED MaxMemPerNode=UNLIMITED

Alternatively, with sinfo under the TIMELIMIT column:

$ sinfo                                                     
PARTITION  AVAIL  TIMELIMIT  NODES  STATE NODELIST               
single        up 7-00:00:00      1 drain* cn097                  
single        up 7-00:00:00      1  maint cn002                  
single        up 7-00:00:00      4 drain* cn[001,061,133,154]    
...