Introduction

Access to all KASK laboratory resources (servers, workstations, cluster, services etc.) requires a single account. Each user has his or her profile available from all computers. Accounts are not created automatically; they are issued ONLY on a student's request for Department's specialty students and co-workers! All questions, remarks, requirements, found errors etc. in the lab should be sent to kasklab@eti.pg.edu.pl e-mail address. Please add “[KASKLAB]” (no quotes) prefix to the email subject.

Login and password

The login is composed of letter "s" (lower case) and student's index number. Example: Jan Kowalski, index number 123456:

  • login: s123456

The newly created account receives a temporary random password. The password must be changed after the first login.

The system will force a password change upon your first login to kask.eti.pg.edu.pl. The SSH process will look like this:

ssh username@kask.eti.pg.edu.pl
(username@kask.eti.pg.edu.pl) Password: (enter temporary password)
Password expired. Change your password now.
(username@kask.eti.pg.edu.pl)Current Password: (enter temporary password)
(username@kask.eti.pg.edu.pl) New password: (enter new password)
(username@kask.eti.pg.edu.pl) Retype new password: (re-enter new password)

Note: The ADMan system is no longer active. To reset your password, you need to send an e-mail to the administrator.

User accounts will be deleted after one year of inactivity. After this time, a warning e-mail will be sent. If there is no response within 30 days, the account and all associated data (including the home directory) will be permanently deleted.

User profile directories

User profiles are stored on a remote storage array. All user files are mounted to each machine.

The servers/workstations

The laboratory consists of:

  • 18 workstations (des01 – des18. in domain lab.kask.eti.pg.edu.pl) in room 527,
  • Compute servers (in domain lab.kask.eti.pg.edu.pl): apl10, apl11, apl12, apl13, apl18, apl19, apl20, apl21,
  • Access node: kask.eti.pg.edu.pl (SSH access – DO NOT RUN COMPUTATIONS HERE!),

Remote Access to the Laboratory

Remote access to the laboratory infrastructure is available exclusively via SSH through the primary gateway server (kask.eti.pg.edu.pl). Direct external access to internal workstations and compute nodes is not permitted. You must first establish a connection to the gateway, and from there, initiate a secondary SSH session to your target machine.

Step-by-step connection guide:

  1. Connect to the gateway server:

    ssh username@kask.eti.pg.edu.pl

  2. Connect to internal resources (from the gateway):
    • To access lab 527 workstations (e.g., des01):

      ssh des01

    • To access compute servers (e.g., apl10):

      ssh apl10

Resources Reservation

Resources reservation is mandatory for most of the hosts. Currently mandatory reservation is required for hosts:

  • apl10
  • apl11
  • apl13
  • apl18
  • apl19
  • apl20
  • apl21

It is recommended to make reservations for hosts:

  • apl12

You do not have to reserve apl12, but users with reservations have priority! If you will schedule calculations without making a reservation you will be notified via e-mail. Users not making or respecting reservations will be banned from the GPU servers.

Reservation rules

  1. Students
    1. only GPUs from one host can be reserved at the same time
    2. the reservation can be done for up to 3 days
  2. Faculty members
    1. you can reserve any number of the GPUs
    2. you can reserve any number of hosts
    3. the reservation can be done for up to 7 days

In both cases after finishing calculations please close all unnecessary sessions, screen, tmuxes, etc.

TensorHive system

To make a reservation please go to the following site:

https://kask.eti.pg.edu.pl/tensorhive/

You have to register providing the user name equal to the one in the laboratory. Before hitting the register button you have to append the key displayed there to your .ssh/authorized_keys file on any of the KASK servers, preferably on kask.eti.pg.edu.pl. Without that the registration will fail as this serves as a check that you have a valid KASK account.

Computers in room 527

des01.lab.kask.eti.pg.edu.pl - des18.lab.kask.eti.pg.edu.pl

  • CPU: 13th Gen Intel Core i7-13700K (24 threads)
  • RAM: 30 GiB
  • GPU: NVIDIA GeForce RTX 4070 Ti 12GB

Servers

If local conditions allow the local storage is made available for users. The local storage space is available under /local_storage (and /ssd_local where available). There is common directory writable by all users that can be used to store temporary data.

The /local_storage and /ssd_local directories should be temporary only! Files belonging to users that did not login for extended period of time can be deleted without any warning!

apl10

  • CPU: 2x Intel Xeon Platinum 8558 (192 threads)
  • RAM: 503 GiB
  • GPU: 1x NVIDIA H100 NVL 96GB
  • Storage: local_storage - 1.8 TB

apl11

  • CPU: 2x Intel Xeon Silver 4316 (80 threads)
  • RAM: 251 GiB
  • GPU: 1x NVIDIA RTX A4500 20GB, 1x NVIDIA A100 80GB PCIe
  • Storage: local_storage - 3.5 TB

apl12

  • CPU: 2x Intel Xeon E5-2680 v2 (40 threads)
  • RAM: 125 GiB
  • GPU: 1x NVIDIA GeForce RTX 2080 Ti 11GB
  • Storage: local_storage - 1.8 TB

apl13

  • CPU: Intel Xeon Gold 6248R (48 threads)
  • RAM: 187 GiB
  • GPU: 2x Quadro RTX 8000 48GB
  • Storage: local_storage - 11 TB, ssd_local - 1.8 TB

apl18

  • CPU: AMD EPYC 9745 128-Core (256 threads)
  • RAM: 498 GiB
  • GPU: 1x NVIDIA GeForce RTX 5090 32GB
  • Storage: local_storage - 1.9 TB

apl19

  • CPU: 2x Intel Xeon Silver 4210 (40 threads)
  • RAM: 376 GiB
  • GPU: 8x Quadro RTX 6000 24GB
  • Storage: local_storage - 3.6 TB, ssd_local - 79 GB

apl20

  • CPU: 2x Intel Xeon Silver 4210 (40 threads)
  • RAM: 376 GiB
  • GPU: 4x Quadro RTX 5000 16GB
  • Storage: local_storage - 3.6 TB

apl21

  • CPU: 2x Intel Xeon 6736P (144 threads)
  • RAM: 499 GiB
  • GPU: 2x NVIDIA RTX PRO 6000 Blackwell Server Edition 96GB
  • Storage: local_storage - 1.8 TB

Web deployment

Interesting student projects can be deployed using KASK resources. KASK employees have to confirm that the project should be deployed. The basic deployment environment is Docker Swarm.

If a project is to be deployed, it must provide a proper docker-compose file, and the application image must be available in a container registry (e.g., KASK GitLab Registry). Building images manually from a Dockerfile is not possible on the department servers!

All needed sources, dependencies, and images have to be specified within the docker-compose file and should be pulled automatically. No manual steps should be involved in project deployment!

Project updates should also be automated using the registry and docker-compose.

The applications have to work behind multiple reverse proxy servers! If it needs some special headers set, proper Nginx configuration should also be provided! The Docker Swarm nodes are behind 2 reverse proxy servers with load balancing!

I cannot directly expose ports to the outside world!

Projects management software

There is a Gitlab instance deployed. Every account holder can login and manage his or her projects. Please bear in mind the resources limits. Each user by default can create 2 projects. The site can be found at the following URL: https://kask.eti.pg.edu.pl/gitlab.