Documentation

• 1: About DAIC
• 1.1: Contributors and funding
• 1.2: Advisors and Impact
• 2: Running Jobs
• 2.1: Job Submission
• 2.2: Interactive Jobs
• 3: Data
• 3.1: Storage
• 3.2: Data Transfer
• 4: Software
• 4.1: Modules
• 4.2: Containers
• 5: System
• 5.1: Specifications

1 - About DAIC

What is an HPC cluster?

A high-performance computing (HPC) cluster is a collection of interconnected compute resources (like CPUs, GPUs, memory, and storage) shared among a group of users. These resources work together to perform lengthy and computationally intensive tasks that would be too large or too slow on a single computer. HPC is especially useful for modern scientific computing applications, where datasets are typically large, models are complex, and computations require specialized hardware (such as GPUs or FPGAs).

What is DAIC?

The Delft AI Cluster (DAIC), formerly known as INSY-HPC (or simply “HPC”), is a TU Delft high-performance computing cluster consisting of Linux compute nodes (i.e., servers) with substantial processing power and memory for running large, long, or GPU-enabled jobs.

What started in 2015 as a CS-only cluster has grown to serve researchers across many TU Delft departments. Each expansion has continued to support the needs of computer science and AI research. Today, DAIC nodes are organized into partitions that correspond to the groups contributing those resources. (See Contributing departments and TU Delft clusters comparison.)

1.1 - Contributors and funding

The Delft AI Cluster (DAIC)—formerly known as INSY-HPC or simply HPC—was initiated within the INSY department in 2015. In later phases, resources were merged with the ST department (collectively called CS@Delft) and expanded further with contributions from other departments across multiple faculties.

Contributing departments

The cluster is available only to users from participating departments. Access is arranged through your department’s contact persons (see Access and accounts).

Table 1: Current DAIC-contributing TU Delft departments/faculties

I	Contributor	Faculty	Faculty abbreviation (English/Dutch)
1	3D Geoinformation	Faculty of Architecture and the Built Environment	ABE/BK
2	Architecture
3	Aerospace Structures and Materials	Faculty of Aerospace Engineering	AE/LR
4	Control and Operations
5	Imaging Physics	Faculty of Applied Sciences	AS/TNW
6	Cognitive Robotics	Faculty of Mechanical Engineering	ME
7	Geoscience & Remote Sensing	Faculty of Civil Engineering and Geosciences	CEG/CiTG
8	Intelligent Systems	Faculty of Electrical Engineering, Mathematics & Computer Science	EEMCS/EWI
9	Software Technology
10	Signal Processing Systems, Microelectronics

Funding sources

In addition to funding from contributing departments, DAIC has received support from the following projects and funding sources:

1.2 - Advisors and Impact

Advisory board

Citation and Acknowledgement

To help demonstrate the impact of DAIC, we ask that you both cite and acknowledge DAIC in your scientific publications. Please use one of the following formats:

• Citation:
• APA
• BibTeX
• RIS(EndNote/RefMan)

@misc{DAIC,
  author = {{Delft AI Cluster (DAIC)}},
  title = {The Delft AI Cluster (DAIC), RRID:SCR_025091},
  year = {2024},
  doi = {10.4233/rrid:scr_025091},
  url = {https://daic.tudelft.nl/}
}

TY  - DATA
T1  - The Delft AI Cluster (DAIC), RRID:SCR_025091
UR  - https://doi.org/10.4233/rrid:scr_025091
PB  - TU Delft
PY  - 2024

Scientific impact in numbers

Since 2015, DAIC has facilitated more than 2,000 scientific outputs from participating departments:

These outputs span a wide range of research areas. Title analysis highlights frequent use of terms related to data analysis and machine learning:

Publications using DAIC

2 - Running Jobs

2.1 - Job Submission

2.2 - Interactive Jobs

3 - Data

3.1 - Storage

3.2 - Data Transfer

4 - Software

4.1 - Modules

4.2 - Containers

5 - System

5.1 - Specifications