DAFNI Newsletter - January 2020
The release of DAFNI 1.0 is an important milestone in the progress of the Facility. This release has been two years in the making, and marks a pivotal moment in the DAFNI calendar. Our conference at The Royal Society in June 2019 was a great success and hosted keynote speakers from across the UK’s infrastructure community, including Sir John Armitt, Chair of the National Infrastructure Commission.
Whilst celebrating the achievements thus far, we are also actively working on further development stages of the facility, with the help of you, our partners on this journey. Our datasets are growing in number and volume, and we are very much interested in entering into discussions with the community to bring models and data onto the platform.
We'll send out subsequent newsletters every month, with this being a first 'bumper' issue!
Get involved: If you are interested in collaborating with us on a pilot project, or hosting datasets on the DAFNI platform, please get in touch.
Marion Samler, DAFNI Partnership Manager
New Project Lead for DAFNI
We are also very pleased to announce that Dr Brian Matthews is joining the DAFNI team as our new Project Lead. Some of you will know Brian as he was involved in the early work to secure funding for the DAFNI Facility. Brian will lead the DAFNI team to develop data and modelling infrastructure to support research into national infrastructure. He has over 30 years of experience in R&D development in computing, with a focus on tools, methods and standards for managing accessing research data from scientific experiments. He took a leading role in the development of the data management infrastructure that supports the ISIS Neutron and Diamond Light Sources, and has worked extensively on European programmes on Data Infrastructures. He is co-investigator and technical lead on the Physical Sciences Data-science Service, one of EPSRC’s National Research Facilities.
Thank you to Dr Peter Oliver for his sterling work not only stepping in to Project Lead the facility but his continued support as a sponsor for DAFNI.
DAFNI functionality and releases
Development of DAFNI continues apace with agile sprints ensuring that features are added regularly. In September we welcomed our first external users onto DAFNI and allowed them to add private datasets and download public datasets. During the last quarter we’ve added improvements to the workflow creation page, the ability to share datasets so that they are visible but not usable, the possibility for users to request a change of visibility in the data they upload, as well as various bug fixes and UI optimisations.
Version 1.5 of DAFNI will launch in February with further functionality added – to be covered in next month’s newsletter!
We also continue to add datasets to DAFNI in areas from climatic and meteorological to geospatial, economic infrastructure data, demographics and social infrastructure data such as housing valuation and building stock data.
UKRI announces e-infrastructure roadmap
We’re very pleased that DAFNI, described as a “scalable hybrid cloud development supporting storage and querying of highly heterogenous datasets in a multi-modal architecture..” is included in the [UKRI’s Research & Innovation Infrastructure Roadmap - aka Infraportal. Access the UKRI report “The UK’s research and innovation infrastructure: opportunities to grow our capability” and find out more about key e-infrastructure along with needs and opportunities for the UK’s capability.
Datasets on DAFNI
The most recent addition to our data family is the UK 2011 Census Data. This comes from the Office for National Statistics Nomis Database bulk download service. A member of the STFC development team said: “This is a landmark dataset for us, as hosting it on DAFNI enables direct access to the files as they are available from within the DAFNI platform. This means that the user would not have to search and download the information, but could access it directly, and thus speed up the process. This will enhance population-based research and there is great potential to link with models on the DAFNI platform for more efficient and collaborative research.”
Security at the heart of DAFNI
Read our latest insight from Pete Feeney, DAFNI’s security engineer, where he explains why and how security is enshrined in everything we do at DAFNI. DAFNI 1.0 offers enough High Performance Computing power to make running complex models, across a variety of scenarios, a possibility in minutes rather than days. With the sheer amount of data DAFNI handles, coupled with complex ownership issues, data sharing can be hard to achieve for a variety of reasons, from security to licensing. Achieving a balance between a system that’s secure yet accessible is our ongoing challenge. Read more
Piloting into the future
Become a pilot
Pilots have really influenced how we have created the DAFNI Facility so far and are a crucial aspect to ongoing development. We are actively seeking new pilots to bring new models and data onto the platform, especially those that ‘stress test’ functionality i.e. GPU, CPU, RAM and storage, big databases. If you are interested in working with us there is no cost involved, all we ask is that you complete our Google form, are free to work alongside our software engineers and thereafter committed to working with DAFNI on an ongoing basis. If you would like to discuss, please contact Marion Samler, DAFNI Partnership Manager.
Pilots in progress
DAFNI Pilot projects have soared within the last six months, and we have been working closely with researchers to bring their models to life within DAFNI. The two most recent projects to join the team, are SPENSER and a Railway Station Demand Model.
SPENSER, a Population Forecast Model, developed by Andrew Smith and Nik Lomax Leeds University, provides a collection of tools for the modelling of the UK population on both Local Authority and national levels, in terms of future growth, projected migration and household structure. This helps predict accurately where infrastructure investments are needed. Read more
Rail station demand modelling
The Railway Station Demand Model, developed by Marcus Young and Simon Blainey at Southampton University, generates a demand forecast (predicted trips per year) for one or more proposed local railway stations. If required, it can also produce an analysis of the potential number of passengers who would change to a new station(s) and what net impact a new station would have on rail use. Read more
Benefits of using DAFNI
- Easy remote access via new web-based interface: No need to install software on the client side.
- Ease of use: The models are accessible to non-experts as no prior knowledge is needed
- Faster model run-time: DAFNI hardware has cut this considerably.
- Web-based visualisation: Allows for detailed comparison of results
- Scenario manipulation: Users can experiment with and manipulate scenarios to their own requirements, thus being able to look at a wider range of situations.
- Map-based visualisation of the results for the user to gain a better understanding of the impact on the surrounding area.
- Links to other DAFNI models: Hosting models on the DAFNI platform will enable them to be linked with other models as the platform grows, thereby improving the efficiency of nationwide infrastructure research on a more holistic level.
DAFNI in the news
Recent press coverage for DAFNI includes:
- Government & Public Sector Journal – November 2019 – introducing DAFNI and why it's vital to infrastructure projects in the UK and beyond.
- Civil Engineering Surveyor – October 2019 – how DAFNI 1.0 is already in Beta testing with eight organisations across the UK
- UK Construction Online – August 2019 – highlighting how DAFNI can change the face of infrastructure research by providing a way to map and model different infrastructures, such as transport links, water and sewerage and power supplies, and their interdependencies on one platform.
- Civil Engineering – August 2019 – the benefits of DAFNI to civil engineers.
- The Civil Engineer blog - June 2019 - how research and computing power can help deliver more resilient, efficient & environmentally appropriate infrastructure.
The STFC DAFNI Team is delivering a series of seminars to universities to provide more information about DAFNI and how it could be used in infrastructure research, to encourage participation and involvement in DAFNI. Participating universities include Bristol, Cranfield, Sheffield and UCL. These sessions have been well received by researchers at Universities and have resulted in follow-on collaborations. To request DAFNI to come to your University, contact Marion Samler, DAFNI Partnership Manager.