NRP: National eScience Centre Research Platform

Lead Research Organisation: University of Edinburgh
Department Name: Sch of Informatics

Abstract

The core funding from the e-Science Core Programme for the National e-Science Centre (NeSC) has enabled us to be leaders in the development of e-Science methods and technology to facilitate research in all disciplines. Our own research has a particular focus / improving techniques for exploiting the ever-growing wealth of distributed scientific data and knowledge. This data may be obtained through experiment, simulation, specialised sensors and human interaction. Exploitation involves every step from obtaining and indexing data to using it to test hypotheses or discover patterns in order to create knowledge. Our work enables researchers in a wide variety of disciplines to be more productive as they exploit data that is growing in scale and complexity.This research is complemented by engaging in intensive collaboration with application domains within our universities, in Scotland, in the UK and internationally to drive forward our understanding, methods, algorithms, architectures and technologies. We then integrate the results and provide services to a wider community through channels such as NeSC workshops, eSI, OMII-UK, EGEE, NextGRID, OGF, Grid Computing Now! and joint research publications. The case for support includes several illustrations of existing collaborations which we will continue to develop further.By providing core support and flexibility, this platform grant will enable us to continue and extend our strategic programme and collaborative relationships. We will continue to undertake research into each contributing domain through joint grants that address explicitly identified technical challenges. Through past experience we recognise that this is often facilitated through initial engagement and prototyping work with those domains. The resources provided by this platform grant will support these preparatory explorations. The combined knowledge and engagement of the named staff and investigators on the platform team will enable us to exploit synergies between our collaborative projects, which would not otherwise be feasible. The continuity provided by the platform grant will enable us to commit to five year's collaboration with our science partners. The case for support provides many illustrations of collaborative research projects currently in development, based on existing work and confirmed by letters of support. Our research strategy extends beyond these collaborations to interdisciplinary and fundamental research that will benefit many disciplines at a number of geographic scales: within the two universities, across Scotland, throughout the UK and internationally.The research will be steered by the investigators with feedback and advice from regular meetings of an applications forum. The NeSC's support for application communities will draw on the rich research environment in the universities and in Scotland. Our support will be delivered through easy-to-use high-level tools operating over an ecosystem of dependable and flexible resources. These resources may be provided locally, by research organisations and in the internet cloud .An incremental, pragmatic and empirical approach is necessary to ensure that we meet researchers' real needs. We will spot opportunities to exploit data or computation for specific goals then integrate the successes to provide an increasingly productive research environment for multiple application domains. The NeSC research programme must deliver what researchers require and adroitly adapt the supporting systems to obtain synergy, sharing, dependability and precision. Sustained engagement significantly increases the power of this approach through refinement and cross-fertilisation between the different applications.

Publications

10 25 50
 
Description The eScience community developed expertise in collaborating between disciplines on data-driven methods, working with both observed data and results from simulations. They developed techniques for presenting the information in ways that assist decision making. With NRP these developed into a collaboration on ischemic stroke brain imaging with Prof Joanna Wardlaw at the Western General hospital funded by the Scottish Funding Council. It developed as a collaboration with Prof Jason Swedlow in Dundee University in a BBSRC-funded project on processing biomedical microscopy images. It developed into a project collaborating on Rock Physics with Prof Ian Main at Edinburgh School of Geosciences, and Prof Philip Meredith at UCL, funded by NERC. This in turn led to subsequent collaborations in FP7 projects ADMIRE, VERCE on computational seismology and ENVRI, and H2020 project ENVRIplus on 23 environmental science research infrastructures, including those endorsed by ESFRI. This has developed into a new H2020 project DARE, Delivering Agile Research Excellence, which is acelearting paths between research development and production as well as pushing the limits of data scale, computational scale and complexity. We are now working with the Farr Institute on medical image research. It has recently led to further collaborations with geoscientists, in the area of using smart phones as seismometers and of integrating data from many sources to support improved emergency response after earthquakes and better build back, in part supported by the NERC REAR project, but also the topic of bids led by Edinburgh's School of Geoscience, which have brought in a DTC and an Expert Centre hub from NERC (Atkinson is actively engaged but not formally involved except as School of Informatics representative. The scalable abstraction of data-streaming underpinning the NRP's technology has proved to have sustainable value because of simplicity, presentation as a python library and the ability to optimally use multiple classes of target platform with changes of scripts or semantics. The separation between conceptual models and implementation models is proving very effective in several contexts, particularly the ESFRI EPOS.
Exploitation Route Joint projects as listed above. The individuals and their students engaged in many of the data science and digital society activities. Many of the Edinburgh staff now work in fields like data science, particularly Dr Jano van Hemert who has a leading research role in Optos, pioneering opthalmology technologies, and many of the PhD students are now in companies including a UK cloud company and a US social media company. The deputy director of NeSC who was much involved in the NRP bid played a leading role at CERN LHC thereafter. One of the PhD students nurtured in the NRP context, Dr Fan Zhu, now plays a leading role in autonomous vehicle systems at BAIDU, where he is Senior Software Engineer in the IDG (Intelligent Driving Group).
Sectors Digital/Communication/Information Technologies (including Software),Education,Environment,Healthcare

URL http://research.nesc.ac.uk
 
Description The findings have been used in education (mainly tertiary) and in application fields by the members of the eScience community and their students. The ideas, methods and technology developed during NRP have fed into many aspects of subsequent transdisciplinary and interdisciplinary applications of data-driven methods, in subsequent projects. These include: 1. The European FP7 VERCE project where scientific workflows for computational seismologists warranted the development of an abstract data-streaming technology dispel4py. This also stimulated work on provenance-data capture and use and on conceptually structured catalogues. The data-streaming technology was continued by national geo-science centres, HPC centres and by the EOSC pilot. It is now a key element supporting climate-impact modelling and EPOS seismologist support by the DARE project (see below), The PROV work has been adopted by the ENVRI community (see below) and EPOS, as well as being a key component of DARE. The Conceptual Catalogue work has been incorporated in the W3C DCAT standard and is in use for the EPOS ESFRI. 2. It led to the award of the ENVRI project & H2020 ENVRIplus now influencing 23 Environmental Research Infrastructures. This included the recruitment of Dr Paul Martin by University of Amsterdam as he had contributed to the scientif workflow framework and the generic e-Research Infrastructure reference model with its ontology, OIL-E. and 3. H2020 DAREworking with computational seismologists and climate impact modellers using the dispel4py abstraction to run rapid assessments of earthquakes to inform responders. The architectures and technology are used by EPCC in the University of Edinburgh and in their consultancy work. Local companies have been influenced by staff who develop experience in the NRP context, in particular 1. Drs Jano van Hemert and Michelle Galea are leading Optos research and Optos-university relationships respectively. 2. Gary McGilvary is a product manager at Cloudreach. Further afield: 1. Dr Chee Sun Liew leads Malasian support for scientific and distributed computation as well as leading his research group, 2. The visit of Dr Rosa Filguera to ISI in University of Southern California has established collaborative use of NRP results, including dispel4py. 3. Dr Rosa Filguera influenced BGS's data-intensive strategy while she was a senior data scientist there. 4. Now Dr Rosa Filguera, as a Senior Data Scientist in EPCC, is helping develop approaches at ATI and in life sciences that build on NRP results. 5. Our collaboration with Dr Ian Overton on analysiing 2500 human genomes to spot similar pathways has now moved to QUB, where he is a reader building up related research. We continue research into federating data-intensive resources from autonomous organisations to meet environmental research and geo-hazard emergency response requirements. This will feed into a collaborative DTC with geosciences in Edinburgh and feed into a collaboration recently award five-years funding for a hub. We have contributed to the development of safe havens for medical images. We will continue to improve the engineering and conceptual framework for data-streamimng scientific methods encoded as dispel4py workflows to achieve computational efficiency and to support discipline crossing research behaviours.
First Year Of Impact 2013
Sector Digital/Communication/Information Technologies (including Software),Environment,Healthcare
Impact Types Cultural,Societal,Economic

 
Description BBSRC Grouped
Amount £350,064 (GBP)
Funding ID BB/I000755/1 
Organisation Biotechnology and Biological Sciences Research Council (BBSRC) 
Sector Public
Country United Kingdom
Start  
 
Description Horizon 2020 Collaborative Project
Amount € 14,998,034 (EUR)
Funding ID 654182 
Organisation European Commission 
Sector Public
Country European Union (EU)
Start 05/2015 
End 04/2019
 
Description Horizon 2020 Research Infrastructure
Amount € 2,957,500 (EUR)
Funding ID 777413 
Organisation European Commission 
Sector Public
Country European Union (EU)
Start 01/2018 
End 12/2020