Chalmers Open Digital Repository

Välkommen till Chalmers öppna digitala arkiv!

Här hittar du:

  • Studentarbeten utgivna på lärosätet, såväl kandidatarbeten som examensarbeten på grund- och masternivå
  • Digitala specialsamlingar, som t ex Chalmers modellkammare
  • Utvalda projektrapporter

Forskningspublikationer, rapporter och avhandlingar hittar du i research.chalmers.se

Enheter i Chalmers ODR

Välj en enhet för att se alla samlingar.

Senast publicerade

  • Bild (thumbnail)
    Beyond the Data Center: Distributed Computing on a Raspberry Pi 5 Cluster
    (2026) Borg, Livia; Burman, Emil; Forsberg, Axel; Fredriksson, Mathias; Tiberg, Emily; Westman, Filip
    Distributed computing clusters are commonly used to provide scalable computation and large memory capacity for demanding workloads. In recent years, single-board computers have become increasingly capable and power-efficient, making them an attractive low-cost alternative for building small-scale distributed systems. How ever, creating such clusters in a way that is scalable, practical, and user-friendly remains challenging due to limited hardware resources and the need for lightweight management and monitoring solutions. This thesis investigates how a distributed computing cluster built from single-board computers can be made practical through lightweight orchestration and purpose-built observability tooling. A central contribution is a custom telemetry system designed for resource-constrained nodes, where existing monitoring solutions impose unnecessary I/O on storage-limited hardware and offer limited control over which metrics are collected and how frequently they are reported. The system collects, transmits and visualizes hardware and performance metrics in real time through a custom web based interface while imposing no measurable impact on workload performance. To evaluate the system, a Raspberry Pi 5 cluster was constructed using Kubernetes for orchestration. Three workloads were deployed to stress different dimensions of the cluster: matrix multiplication for parallel compute throughput, distributed pass word recovery for CPU-intensive data parallelism, and split large language model inference for distributed memory capacity. The results show that the cluster achieved significant performance improvements compared to single-node execution, particularly for highly parallelizable workloads. The system also demonstrated good power efficiency and highlighted the advantages of distributed memory for running larger LLMs. However, the limited computational performance of individual Raspberry Pi nodes means that many devices are required to approach the performance of a conventional high-performance machine. Overall, the work demonstrates that single-board computer clusters can provide a flexible and energy-efficient platform for distributed computing, especially when combined with lightweight orchestration and observability tools.
  • Bild (thumbnail)
    A Monitoring System for Laboratory Workstations
    (2026) Gralén, Nils; Karlsteen, Albin; Menor Löwegren, Nicholas; Edvin, Palmqvist; Paulsson, Elof
    University computer laboratories rely on a large number of shared workstations that must remain operational and available to students. However, hardware-related incidents, such as component theft, unexpected shutdowns, and network disconnections, can be difficult to detect using IT monitoring solutions. The proposed monitoring system was implemented as a centralised client-server architecture consisting of lightweight monitoring clients, a central server, a persistent database, and a graphical user interface for administrative oversight. The system combines heartbeat monitoring, hardware validation, chassis intrusion detection, and automated recovery mechanisms to provide centralised situational awareness and alarm handling. The system was evaluated through passive testing and active threat simulations performed on laboratory computers at Chalmers University of Technology. The evaluation demonstrated successful detection of simulated tampering events, stable heartbeat communication, low resource utilization, and reliable alarm handling with minimal interference with normal laboratory usage. Although the system remains a proof of concept prototype, the project demonstrates that event-driven hardware monitoring can provide an effective foundation for improving hardware security and operational awareness in shared university computer laboratory environments.
  • Bild (thumbnail)
    A Monitoring and Alert System for Shared Computer Laboratories
    (2026) Akar, Sukaina; Hagberg, Angelika; Kadamani, Houmam; Moberg, Joel; Tariq, Sufian
    Shared computer laboratories at universities provide essential resources to students, but they are also vulnerable to theft of internal hardware components, such as graphics cards, memory modules, and storage devices. Detecting such thefts is challenging because many benign events, including maintenance, configuration changes, and temporary network issues, can produce symptoms similar to physical tampering. This thesis presents the design, implementation, and evaluation of a prototype software-based monitoring and alert system for shared computer workstations. The system uses a client-server architecture in which lightweight clients installed on lab oratory computers send periodic heartbeats and provide information to the server. The server applies context-aware alarm logic to classify events and reduce false alarms. The system was evaluated through scenario-based testing in a laboratory environment. The results show that the system reliably detects hardware component changes and that the context-aware alarm logic reduces false alarms compared to an approach in which every unavailability event triggers an alert. The system was able to distinguish between benign and suspicious events in evaluated scenarios where contextual evidence was available. However, certain situations, such as a single machine losing power, remain inherently ambiguous, as they produce system-level signals identical to those caused by physical tampering.
  • Bild (thumbnail)
  • Bild (thumbnail)
    Hur alkalinitet i cement påverkar biofiber över tid
    (2026) Brage, David; Elfving, Emilia; Eliasson, Hanna; Friman, Madeleine; Nikq, Valdrin
    The cement industry contributes significantly to global carbon dioxide emissions and drives the need for sustainable alternatives such as biofiber reinforcement in cement. The highly alkaline environment of cement has the potential to degrade biofibers, which impairs their long term performance. This study investigates the microscopic effects of the alkaline cement environment on biofibers coated with ex foliated graphite over time. Fibers from sugarcane and corrugated paper were coated with exfoliated graphite using two distinct methods. The fibers were exposed to an alkaline solution for 1, 7 and 28 days. The samples were analyzed using FTIR, XRD, SEM, EDS and TGA. Results indicate that the exfoliated graphite coating had an impact on the surface interaction between the biofiber and the alkaline solution by affecting the deposition of Calcium (Ca) on the biofiber surface. Additionally the thermal stability of the biofiber differed depending on whether the coating was present after conditioning for 28 days. One of the methods yielded a significantly better coating and higher chemical stability than the other method. Some structu ral degradation may still have occurred depending on the fiber type and application method. The coated biofibers demonstrated better resistance and stability compared to uncoated reference samples. Exfoliated graphite coating provides a potential for protecting biofibers in an alkaline environment like cement. Future research must address long term exposure and process upscaling before industrial implementation.