TRE Satellite Event notes - Open source infrastructure and collaborations#


  • How many open-source TREs do we want?

  • Do we need separate collaborations for each open-source TRE?

  • Can we collaborate across different TRE platforms?


What are people currently using?#

  • On-prem

  • Azure

  • AWS

What open-source solutions exist?#

  • TREEHOOSE (used at Dundee)

    • based on AWS service bench

  • Turing Data Safe Haven

    • BSD licence

  • Microsoft AzureTRE

    • MIT licence

Who is able to work on a common codebase?#

  • Azure

    • Turing

  • AWS

    • Dundee (HIC)

  • If a single code base is too difficult to work in practice, it would still be very valuable to have some common governance, policy, guidelines, specifications.

Who is interested in deploying and evaluating multiple TRE solutions?#

  • Turing

  • UCL

  • Sheffield

  • Dundee

Who is interested in working on common governance specification?#

  • Turing

  • Dundee

  • EPCC


  • On-prem open stack implementations using Terrafrom etc.

  • Trade off between ‘transferability’ and ‘conveniences/maintainability’

    • Can just use VMs, more transferable but adds managment burden

  • Container based solutions .e.g. kubernetes. Another layer of abstraction that removes cloud-vendor dependencies.

    • Running graphical desktops is difficult (maybe impossible for Windows)

How to include HPC?#

  • Burst out for running particular jobs?

Actions/next steps#

  • Common governance - take the initiative to set this up

  • Send invitation to a kick-off meeting

  • Similiar discussion on code - What do we have already, do we want to combine or adopt a single codebase