We were pleased to announce that we received a further two years of funding from EPSRC’s Resource Only Strategic Equipment fund at the DAFNI Conference earlier this year. This allows us to bring the DAFNI platform to a wider group of users and really start to build on the 4-year development roadmap by providing access to the many of the benefits the platform has to offer our community of researchers.
Since the conference, the platform has been continuing to grow, and on the last count DAFNI has 1,876 datasets, 176 models and 1,474 user workflows, as part of an active user community who are using DAFNI to develop their research.
If you are a researcher and would like to become part of the community of DAFNI users, we would welcome your application; full details of how to access DAFNI are shared below.
We are continuing our programme of DAFNI roadshow events, a new schedule will be released soon, to explore how DAFNI can support new researchers. We are also establishing a training programme as if you would like to attend one our training events (dates below) if you want a deep dive into how to use DAFNI. Keep an eye out for the dates of these events – we would love to see you there!
Dr Brian Matthews, DAFNI Project Lead
Contents
- Apply for a DAFNI account
- Event news
- Outputs from our July conference
- Recent DAFNI enhancements: DAFNI version 3
- News from our central team
Apply for a DAFNI account
Would you like to apply for a DAFNI account?
We have a ‘light-weight’ application account to gain access to DAFNI. To request your account, please go to our DAFNI website.
Events news
DAFNI technical training
Following a successful training event, we are now launching a series of technical training events for DAFNI.
- 1st December 2021
- 19th January 2022
- 2nd March 2022
- 20th April 2022
- 15th June 2022
- 20th July 2022
If you would like to register your interest in attending any of these dates, please email Rocio at info@dafni.ac.uk. To attend the event you will need to have experience of entering code through a command-line interface, for more information please refer to our booking page as there are pre-requisites to attend the training https://www.eventbrite.co.uk/o/dafni-31793198351.
DAFNI roadshows events
DAFNI Roadshow events continue. Join us on 22nd November to hear from Dr Juste Raimbault, a DAFNI Champion, about OpenMole and how this is now integrated on the DAFNI platform
At his event you will also hear from:
Denise Pumain who is an Honorary Professor at Université Paris who has developed an integrative approach to urban systems dynamics, combining an evolutionary theory for systems of cities with the development of simulation models and harmonised databases.
Romain Reuillon, a researcher at Complex Systems Institute in Paris (ISC-PIF) and is the founder of the OpenMOLE project, and one of its main developers for more than 10 years. The focus of his research is distributed computing and the validation of simulation models, in an interdisciplinary context
View the videos from our July conference
We captured the event and the videos and presentations are available on our website.
- Professor Dame Ottoline Leyser // UKRI Chief Executive
- Professor Jim Hall // Chair of DAFNI Governance Board, University of Oxford
- Professor Gordon Masterton // Chair of DAFNI Governance Board, University of Oxford
- Dr Brian Matthews // DAFNI Project Lead, STFC
- Kathryn Brown // DAFNI Project Lead, STFC
- Mark Gaskarth // Head of Digital Twins at EPSRC
- Professor Liz Varga // Head of UCL’s Infrastructure Systems Institute
- Professor David Wallom // UK Centre for Greening Finance and Investment, University of Oxford
- Professor Robert Nicholls // Director, Tyndall Centre, University of East Anglia
- Professor Daniel Coca // Head of ACSE Department, University of Sheffield
Latest developments on the DAFNI platform
DAFNI launches version 3
- Added a publisher field to Model and Workflow metadata to allow users to enter a publisher rather than showing a UUID
- Added versioning to Workflows
- Users with edit permissions on a Workflow will now be able to see an edit button on the Workflow details page which will allow the user to edit the metadata and steps of the Workflow
- To support Workflow versioning, users now need to set a version message for Workflow versions
- Version tags have been added to Workflow versions, three tags are available – Latest which is automatically assigned, Default and Deprecated. Users with edit permissions will be able to assign a single Workflow version as the default version of their Workflow
- Workflow Catalogue prioritises displaying Default versions of a Workflow, it will fall back to the Latest version of the Workflow when no Default is specified.
- Added version bar to Workflow version details page to allow users to browse the version history of a Workflow as well as convenient buttons to take you straight to the Default or Latest versions of a Workflow
- Added Parameter Sets functionality – this is a large overhaul of the way the Workflow system works and is the reason why we have made this version of DAFNI a new major version
- Model Steps in Workflows no longer accept parameter values or datasets, these are now provided in Parameter Sets
- When creating a Workflow for the first time you will now need to specify: Workflow metadata, the steps of your Workflow, metadata for the first Parameter Set of your Workflow and the values/datasets for that Parameter Set
- Alternatively, when specifying the metadata for a Parameter Set you can select to create a Parameter Set with default values. However, if your Workflow contains uses a Model that has a required parameter that doesn’t have a default value provided you won’t be able to create a default Parameter Set
- The Workflow details page now shows a list of Parameter Sets that have been created for the Workflow, in this section there is a button to allow you to create a new Parameter Set
- Next to each Parameter Set in the list are two buttons: an edit button (editing a Parameter Set will overwrite it) and an execute button
- To execute a Workflow, a Workflow version and a Parameter Set are now required. The Workflow Version must be saved in the DAFNI catalogue but the Parameter Set does not have to be saved. This is to allow Workflow Instances to be linked to a Workflow version
- Workflow Instances are now displayed on the Workflow details page, clicking on the Instance in the list will take you to the status page for the Workflow Instance
- Unfortunately, due to the way Workflow Instances used to be stored, we have been unable to match up Instances that were run before this update with Workflow versions. They are still visible on a new page, that can be navigated to through the Workflows catalogue. In a future update we will add functionality to allow you to link historic Workflow Instances up to Workflow versions
- New documentation pages have been added to explain how to create a Workflow and Parameter Set in this new system
- A YouTube video has also been released to explain the changes that have happened in this release and goes through how to use Workflows and Parameter Sets
- Added an info box explaining how globbing of files works for publisher and visualisation steps of Workflows
- Updated Workflow and Model definitions to use the same casing standard for all fields (snake case)
- You can now search for a Dataset or Version ID in the Data catalogue or when searching for a Dataset to use in a model step of a Workflow
- Reworked the dialog that appears when executing a Workflow so that there is no delay before it opens to reduce confusion around whether the Workflow has been executed and to prevent repeat executions
- The DAFNI web app will now make sure that the step names within a Workflow are unique as if a Workflow contains more than one step with the same name it will fail to execute
- Removed tags and type from Model and Workflow metadata while we re-think how to best use these fields in the future
News from our central team
Climate Resilience Demonstrator
DAFNI is proud to provide the platform to support CDBB’s CReDo project. The Climate Resilience Demonstrator from the National Digital Twin programme seeks to improve resilience across infrastructure networks.
Watch a video released for COP26 which highlights the project: https://youtu.be/iluoK6iKrxE.
Find out more about the CReDO project.
DAFNI in DCMS film for COP26
DAFNI appears in the Department of Culture, Media and Sport (DCMS) video produced for COP26.
DAFNI sets sights on environment and digital twins
Professor Jim Hall was recently interviewed about DAFNI by Mark Say for UK Authority, an online publication which explores and showcases best practice and innovation in the use of technology, digital and data for the delivery of modern public services.
DAFNI on podcast
Why use DAFNI? Find out by listening to Brian Matthews who was recently interviewed about DAFNI during a STFC Data Verse podcast…
SES celebrates DAFNI’s work
The Science and Engineering South community, a hub of world-leading universities, celebrates the work of the DAFNI champions.
DAFNI Immersive Data Space at Bristol
Introducing the DAFNI Immersive Data Space, a new UKCRIC touring immersive space for visualisation of infrastructure data at the University of Bristol. This facility was funded through DAFNI as part of the UKCRIC’s EPSRC research fund.
Open Data Institute seeks collaboration on how the built environment sector can use data to tackle climate change.
