Our research data services at one glance

Inform yourself about the distinct data storage solutions and their specifications



Nextcloud

Nextcloud can be deployed as OnPrem cloud storage in two variants: 

  • Nextcloud intern exclusively for internal use or access via VPN, note that data sharing with external project partners is not possible.
  • Nextcloud extern allows internal and external access. This concept is suitable for data sharing with external project partners.

Basic information:

  • Storage capacity: 10 TB per structural unit
  • Guaranteed storage provision after project completion: 10 years
  • Backup existing

Advantages:

  • Archiv concept
  • Use of OnlyOffice possible

Disadvantages:

  • Cannot be used for specific data analysis
Git Lab

GitLab based on Git depicts a web application which is used for version management of projects, software and data.

Basic informationen:

  • Storage capacity: project-dependent assessment
  • Guaranteed storage provision after project completion: 10 years
  • Backup existing

Advantages:

  • Overview of versions
  • FAIR data storage and provisioning 
  • README for project details
  • Good rights-management
  • Project folder link can be released for publications
  • Commands via user interface or console 

Disadvantage:

  • Cannot be used for specific data analysis
  • Low storage capacity
  • Upload of single files or a ZIP compressed folder at the web interface, otherwise use of specific software such as Sourcetree


Virtual Machine

Virtual Machines (VMs) can be used for efficient data analysis and data storage. Acces to VMs can be achieved via Nextcloud, data are directly provided in the analysis environment. It can be provided for internal use.

Basic information:

  • Storage capacity: 10 TB per structural unit (more is possible)
  • Guaranteed storage provision after project completion: 10 years
  • Backup existing

Advantages:

  • Operating system exists (Linux)
  • Data analysis and data storage
  • Access via remote desktop client 

Disadvantages:

  • Can only be used for web based analysis


Linux Container

Linux Container (LXC) can be used for efficient data analysis and data storage. Acces to LXCs can be achieved via Nextcloud, data are directly provided in the analysis environment. It can be provided for internal use.

Basic information:

  • Storage capacity: 10 TB per structural unit (more is possible)
  • Guaranteed storage provision after project completion: 10 years
  • Backup existing

Advantages:

  • Data analysis and data storage
  • More computing power available

Disadvantages:

  • No complete operating system existing (mock operating system)
  • Commands only via console 
  • Access only via SSH


XNAT

XNAT can be provided for internal or external usage. It is an open-source platform for the administration of image data. 

  • Basic information:

    • Storage capacity: 10 TB per structural unit (more is possible)
    • Guaranteed storage provision after project completion: 10 years
    • Backup existing
    • specific for the administration of image data
DIC-AI-Tool

The DIC-AI-Tool is basesd on the open-source framework Ollama. Different Large Language Models can be used:

  • Ollama3.2 Vision: for general questions, translations, analysis 
  • Deepseek-coder-v2: optimized to improve developer code or to write complete programs 

The use of the AI-Tool hosted by the DIC is approved by the Information Security. When registering for the use, please  read the user information carefully.


Hier klicken

Jupyter Hub

Jupyter Hub can be provided for internal or external usage. 

What can I use Jupyter Hub for?

Jupyter Hub is a versatile tool that is ideal for the following use cases:

  1. Interactive Data Analysis and Exploration: Perform data cleaning, transformation, and exploratory data analysis in a step-by-step and documented manner.

  2. Prototyping and Development: Quickly test code snippets, algorithms, and mathematical models in an immediate feedback environment.

  3. Scientific Computing and Simulations: Run complex computations and simulations in multiple programming languages (e.g., Python, R, Julia) thanks to the kernel architecture.

  4. Reporting and Documentation: Create dynamic reports that include code, results, and explanatory text, ensuring the reproducibility of your work.

  5. Teaching and Training: Serves as a standardized and accessible environment for training in data science, programming, or machine learning.


  • Keine Stichwörter

2 Kommentare

  1. Tim Herrmann sagt:

    Liebe Alice Grünig , Ich hab mal die Seiten angepasst. Unser größtes Manko ist derzeit dass wir keinerlei Möglichkeiten haben im Intranet Speicher als Netzlaufwerk den Klienten anzubieten. Da kommen wir bisher nicht weiter mit dem ITMT.

    1. Alice Grünig sagt:

      Danke dir! Ja, das stimmt wohl. Aber so haben wir zumindest schon mal ein paar Tools bei der Hand.