Suche
Mein Konto
IT disruption at TU Ilmenau: Systems back online after a short break!
Disturbances in the IT network of the TU Ilmenau affect systems at the University of Erfurt. Affected users are asked to be patient.
IT disruption at TU Ilmenau: Systems back online after a short break!
On July 3, 2025, the University Computing and Media Center (URMZ) at the Ilmenau University of Technology reported significant disruptions in the IT network, which also affected the University of Erfurt. The disruptions affected the use of significant systems such as Exchange, Outlook and Moodle, resulting in significant disruption to university operations.
The URMZ announced that the causes of the problems had already been identified and that they were actively working to resolve them. Users of the systems were asked to be patient while data center technicians worked to resolve the problems. The all-clear was finally given at 4:30 p.m.: the systems are fully functional again, and the URMZ thanked the users for their patience.
Technological background
An important part of the technological infrastructure at TU Ilmenau is the GPU Compute Cluster, which was installed in June 2021. This cluster consists of multiple NVIDIA GPUs connected together to dramatically increase computing power. In detail the cluster includes:
- 7 GPU-Server
- 320 CPU-Cores
- 5 Terabyte Hauptspeicher (RAM)
- 36 NVIDIA A100 Tensor-Core GPUs
- 1,44 Terabyte Grafikspeicher insgesamt
- Verbindung über ein 200Gbit/s latenzarmes Netzwerk
By combining these components, the cluster is able to efficiently process large amounts of data, which will be crucial for future research projects at TU Ilmenau. Currently, projects such as “Engineering for Smart Manufacturing (E4SM)”, funded by the Carl Zeiss Foundation, and “ThurAI”, an inter-university project that was started in April 2021, benefit from this state-of-the-art technology.
Cybersecurity challenges
However, the use of such powerful clusters also presents challenges in the area of cybersecurity. According to a report on LinkedIn, increasing demands on data centers require proactive measures to protect sensitive research data and intellectual property. This data may include personally identifiable information (PII) or proprietary algorithms, the loss of which could result in significant reputational damage to the University.
The risks are complex and include, but are not limited to, data breaches, ransomware attacks and the risk of internals being spied on by internal and external actors. To counter these threats, various strategies are essential, including:
- Risikobewertung zur Identifizierung von Bedrohungen und Schwachstellen
- Implementierung von Zugriffssteuerungen, um nur autorisierten Personen Zugriff zu gewähren
- Verschlüsselung von Daten sowohl im Ruhezustand als auch während der Übertragung
- Netzwerksegmentierung zur Isolierung von GPU-Clustern vom Hauptnetzwerk
- Regelmäßige Aktualisierung der Clusterkomponenten zur Minimierung von Schwachstellen
- Entwicklung von Notfallplänen zur schnelleren Reaktion auf Sicherheitsvorfälle
- Schulungen zur Sensibilisierung der Nutzer für Cybersicherheit
- Kontinuierliches Monitoring zur schnellen Erkennung von anomalies
Implementing these measures is critical to ensuring the integrity of academic research and maintaining the trust of stakeholders and the public. Universities must continue to work together to meet these challenges in the digital age.