SE Research Seminar in Secure and Distributed Computing: DPS
Selected Topics in Cloud/Edge/IoT and High Performance Computing

Winter Term 2024/25, LVA 703325

 

Time and Location 

    Wednesdays, 13:45 - 15:15 Uhr (start on Oct. 2)
    location: architecture building, Rechnerraum 26

Instructor
    Prof. T. Fahringer, Distributed and Parallel Systems Group, Institute for Computer Science, University of Innsbruck
    Office hours (ICT building, 2nd floor): Wed. 13 - 14.00,
    Email: Thomas.Fahringer at uibk.ac.at

Contents
In this seminar we will cover advanced topics in the fields of distributed system (Cloud/Edge/IoT) and high performance computing (HPC). Both topics are highly relevant and complement each other. Most available computing devices ranging from IoT devices, edge and fog compute and storage systems and cloud infrastructures are connected by the Internet or wireless communication systems and cloud computers are usually HPC systems. Even fog and edge systems frequently include high performance compute devices including GPUs and accelerator hardware. The primary goal of this seminar is study application development, runtime environments, tools, architectures as well as applications for distributed systems and HPC. For distributed systems we largely focus on the cloud/Edge/IoT continuum including IoT, whereas for HPC we cover the full hardware spectrum ranging from multi-core processors, GPUs, accelerators to modern servers, shared memory and distributed memory parallel systems.

          Topics of interest for this seminar include, but are not limited to:
    • Internet Computing Frontiers: Edge, Fog, Serverless, Lambda, Streaming, etc.
      More decentralized approaches to cloud computing. Edge/fog/mist computing, sensor data streaming and computation on the edges of the network. Function as a Service (Faas), Backend as a Service (BaaS), serverless computing, lambda computing.
    • Architecture, Networking, Data Centers
      Service oriented architectures. Utility computing models. IaaS, PaaS, SaaS, *aaS paradigms. Service composition and orchestration. Micro-datacenter, cloudlet, edge, or fog computing infrastructure. Virtualized hardware: GPUs, tensor processing units, FPGAs.
    • Programming environments, runtime systems, compilers and tools for HPC including GPU/FPGA clusters
    • Storage and I/O Systems
      Distributed storage, cloud storage, Storage as a Service, data locality techniques for in-memory processing, storage in the edge.
    • Programming Models and Runtime Systems
      Programming models, languages, systems and tools/environments. Virtualization, containers, and middleware technologies. Actors, agents, programming decentralized computing systems.
    • Resource Management and Scheduling
      Resource allocation algorithms, profiling, modeling. Cluster, cloud, and internet computing scheduling and meta-scheduling techniques.
    • Performance Modelling and Evaluation
      Performance models. Monitoring and evaluation tools. Analysis of system/application performance.
    • Security, Privacy, and trust
      Cloud security and trust. Access control. Data privacy and integrity. Regulation.
    • Sustainable and Green Computing
      Environment friendly computing ecosystems. Hardware/software/application energy efficiency. Power, cooling and thermal awareness.
    • Applications: Data Science, Artificial Intelligence, Cyber-Physical Systems, etc.
      Applications to real and complex problems in science, engineering, business and society. User studies. Experiences with large-scale deployments systems and data science applications. Distributed AI/Machine Learning on the cloud. Data streaming analytics for intelligent transportation systems.
          If you are working on a project in industry or academia or if the topic of your master thesis fits this seminar theme, then you can also propose your current work as a seminar topic. 

Foils of the Initial Discussion
    (available only after the first date of the seminar)

    Organistation of the Seminar
    Introduction to the Seminar

Goals
    Acquire the ability to study and present new research material. In depth understanding of a selected seminar topic.

Target groups
 Master and PhD students of computer science or related sciences which attended relevant bachelor courses such as "Parallele Programmierung" or "Distributed Systems"or relevant master courses such as "Parallel Systems", "Distributed applications in the edge-cloud continuum", or "Scheduling approaches in distributed systems".

Tasks
Students select a seminar topic for which a seminar thesis (approx. 15 pages) and a presentation (30 minutes + 5 minutes discussion) have to be prepared. Both seminar thesis and presentation slides must submitted via OLAT at least 1 day before the presentation. The same material must be printed out and handed over to the instructor at the day of the presentation.

Grading

    Both seminar thesis, presentation and slides have an impact on the grade. All participants must be present at all seminar days. All documents, slides and presentations must be done in English.

Presentation techniques:

    Find some links to learn about presentation techniques.