EIDF's data science virtual desktop service offers access to virtual machines running on the EIDF local cloud. The EIDF service team is delighted to launch the first version of the data science virtual desktop service, offering Linux virtual machines (VMs) through a "desktop in a browser" virtual desktop interface (VDI).The service is intended for users who need more compute and storage than their local resources allow, but who don't need the full power of a high-performance computing or dedicated machine learning system. The EIDF virtual desktops run on a computing cloud hosted entirely within the University of Edinburgh’s Advanced Computing Facility, the high-performance computing data centre managed by EPCC. Desktop machines can be ordered with a range of pre-installed data science software including Python, R, and JupyterLab. The service also offers a full visual desktop rather than a bare Linux command line, enabling the use of graphical tools such as RStudio or Jupyter Notebooks. Our aim has been to make the service as "self-manageable" as possible. EIDF VMs are arranged in projects. A Principal Investigator (in EIDF-speak, the person responsible for a given project, not necessarily a grant PI) can apply for virtual resources for a DDI-based project through the EIDF Application Portal (see Links below). Once approved, the PI can manage this "resource pool" themselves, creating as many VMs as they need within the overall project quota. PIs can create user accounts and give collaborators access to individual VMs within the project. Users access the service either through the EIDF VDI, a web portal that allows them to choose between the VMs available to them, or directly by ssh (if command-line access is all they need). The virtual desktop service is available to University of Edinburgh researchers and their academic and commercial collaborators working on DDI and data science projects, with a flat-rate overhead per person per year. This one-off charge gives users access to virtual machines and disk space up to several hundred terabytes. There's no limit on the number and size of VMs that can be requested, but we will ask project PIs to justify large resource requests and may suggest the use of alternative EIDF or EPCC compute services if appropriate. Users who need very large memory VMs may, for instance, be better off using the Ultra2 large-memory service (see Links below). The virtual desktop service is already supporting researchers from the Bayes Centre and the Edinburgh Futures Institute. We will regularly expand and extend the service to match the ambitions of the DDI Programme and our users. If you're interested, do get in touch! Links Virtual desktops service catalogue and the EIDF Application Portal. Ultra2 large memory service. Contact us Email us at eidf@epcc.ed.ac.uk Join our mailing list to receive updates about EIDF. This article was published on 2023-11-24
