IT
Arguably the most fundamental part of the research environment from the perspective of my research is well functioning and supported Information Technology (IT). The SoG has an excellent level of IT support. A web server is maintained which allows research information to be readily organised and research outputs to be made available on the World Wide Web. Within the SoG the network is continually improved and in the last year wireless access has become available in many parts of the buildings.
During the RAE2008 reporting period the IT hardware available in the SoG has undergone some major changes. This has been driven by reductions in hardware costs and imporvements in cable networking as much as it has been driven by computational and data demands. Through UK Research Council funding and investment by the SoG a 30 Node Beowulf Super Computer is available. This is a great resource although it should be noted that this can always be improved.
The masters computer lab is another excellent resource of around 30 high specification PCs, half of which are updated on an annual cycle. This resource could be improved by installing Condor to make it into a processing resource comaparable in power to the Beowulf Super Computer.
It would be good for the SoG in collaboration with the White Rose Grid to draw up a Service Level Agreement to make available underutilised resources on the NGS . Indeed the University of Leeds as a whole should be looking to do this with ISS clusters.
The SoG has an Access Grid facility which enables new forms of collaboration, seminars and meetings.
World Wide Web
Owing to the excellent IT support, the SoG and the UoL have well functioning web servers. These web servers are a critically important part of research infrastructure. The provide an excellent level of information for the functioning of SoG and UoL.
Additionally and perhaps as important, research is able to be published online to the global community. This is excellent especially from my research perspective. I can publish information online to promote collaborative research. Web based information has major advantages as an information assimilation, dissemination and communication platform.
Research Centres, Groups, Clusters
CCG
The RAE2008 period has been a difficult time for the Centre for Computational Geography (CCG). The CCG is an interdiciplinary university centre which is and always has been mainly comprised of members of the SoG. The sad early retirement of Stan Openshaw precipitated the difficulties during the RAE2008 period. The SoG has been supportive in providing new IT infrastructure for supporting CCG research. The CCG website is now hosted on the SoG webserver and data backup is organised by SoG. The CCG no longer has the burden of maintaining a network of workstations for computaional reasons owing to the availability of other resources. Problems of moving and porting bespoke software to new platforms and reconfiguring the webservices that the CCG webserver provided and indeed the webpage information itself are only partially resolved.
The CCG is widely recognised (by the international research community, by many in UK academia and by standards bodies that deal with geospatial information) as a body of expertise doing computaional geography research at the frontier of the subject.
MASS
The Multi-Agent Systems and Simulation (MASS) group meet regularly as a reading group and to develop a platform geographical tool. This is an excellent example of a reseach group that mainly operates within the SoG drawing members across the SoG Research Clusters. MASS has regular meetings and draws together information and attempts to pool resources from across the broad academic community to develop synergies, develop new tools, new understanding from a geographic and applied perspective.
SoG Research Organisation
The formation of strong and focussed research clusters is useful. However, it needs to be recognised that there are many more than 5 areas of geographic research in which we in the SoG are involved. Some focussed research sit well within this cluster model, but much research to broader geographic interests and it is great the SoG recognises this at the highest level.
