AI Cloud Service

    Facilitet: Udstyr

    • Placering

      AAU Datacenter

      Aalborg Øst

      9220

      Danmark

    Udstyrsdetaljer

    Beskrivelse

    AI Cloud Service comprises of special infrastrature, software stack, and support activities which all gear toward artificial intelligence (AI) and GPU-based computation.

    The service runs SLURM workload manager for automatic resource allocation and supports container technologies for isolated, flexible and reproducible compute environment. Each compute node is a NVIDIA DGX2 machine which contains 16 Tesla Volta V100 GPUs (32GB RAM each), 1.5TB of RAM, 30TB of NVME (SSD) for scratch space. Total GPU computation power is around 6 PetaFLOPS.

    Please check this page to see how you can get access and use the service: https://www.claaudia.aau.dk/platforms-tools/compute/gpu-cloud-ai/

    For general computation needs that use normal CPUs in form of virtual machines to run desktop applications similar to what you would run on a single workstation, please check out our Compute Service.

    Detaljer

    NavnAI Cloud Service
    Anskaffelsesdato01/12/2018

    Fingerprint Udforsk forskningsområderne, hvor dette udstyr er blevet brugt. Disse etiketter er oprettet på baggrund af de relaterede publikationer. Sammen danner de et unikt fingerprint.

  • Publikation

    • 5 Konferenceartikel i proceeding

    A Dual-arm Collaborative Robot System for the Smart Factories of the Future

    Filtenborg Buhl, J., Grønhøj, R., Kjær Jørgensen, J., Mateus, G., Pinto, D., Krunderup Sørensen, J., Bøgh, S. & Chrysostomou, D., 2019, 29th International Conference on Flexible Automation and Intelligent Manufacturing: FAIM 2019. Elsevier, Bind 38. s. 333-340 8 s. (Procedia Manufacturing).

    Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

    Åben adgang
    Fil
  • 1 Citationer (Scopus)
    21 Downloads (Pure)

    Control of HVAC-systems with Slow Thermodynamic Using Reinforcement Learning

    Blad, C., Koch, S., Ganeswarathas, S., Kallesøe, C. & Bøgh, S., 2019, 29th International Conference on Flexible Automation and Intelligent Manufacturing: FAIM 2019. Elsevier, Bind 38. s. 1308-1315 8 s. (Procedia Manufacturing).

    Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

    Åben adgang
    Fil
  • 24 Downloads (Pure)

    Self-learning Processes in Smart Factories: Deep Reinforcement Learning for Process Control of Robot Brine Injection

    Andersen, R. E., Madsen, S., Barlo, A. B. K., Blegebrønd Johansen, S., Nør, M., Andersen, R. S. & Bøgh, S., 2019, 29th International Conference on Flexible Automation and Intelligent Manufacturing: FAIM 2019. Elsevier, Bind 38. s. 171-177 7 s. (Procedia Manufacturing).

    Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

    Åben adgang
    Fil
  • 1 Citationer (Scopus)
    44 Downloads (Pure)

    Projekter

    • 4 Igangværende
  • Reinforcement Learning Baseret Styring til Gulvvarme Systemer

    Bøgh, S. & Blad, C.

    01/01/201931/12/2021

    Projekter: ProjektForskning

    Fundamentals of Thermal Properties of Oxide Glasses

    Smedskjær, M. M., Sørensen, S. S. & Johra, H.

    01/09/201831/08/2022

    Projekter: ProjektForskning