Portal and core infrastructure

The development of the main AURIN application, referred to as the ‘AURIN Portal’ has been in the hands of the Melbourne eResearch Group (MEG) since 2010.  The team has grown from a handful of programmers in the early years to a large and multi-skilled group of 17.  A relatively exploratory phase until late 2012 identified a range of technology and architectural options.  A ‘data-driven’ approach was taken from early-on in recognition of the diversity of datasets and schemas that would be ingested from various sources.  Most elements of the AURIN technology stack were taken through two development iterations before the majority of components were finalised in 2013 and the first fully-realised beta release went live in May 2013.

Subsequent beta versions have been released in September 2013 and March 2014, and the next release (Beta5) will be in September 2014. There is expected to be one further release in March 2015.

The user-interface to the Portal was the subject of a substantial testing and re-design process in the second quarter of 2014.  The Usability Lab at the University of Melbourne under Frank Vetere was engaged and UI specialists design4use brought in to facilitate. 13 participants, experienced in a practice area related to AURIN’s concerns but not familiar with the Portal, attending 60-90 minute sessions spread over 3 weeks. The sessions firstly tested the existing interface and then mock-ups of amended designs. These mock-ups arose out of many hours of discussion within the team after the first round of testing, aided by involvement and facilitation from design4use. The recommendations for a revamped UI were put through a final round of assessment with the MEG team and the AURIN Office and were written into forward development program. Much of the revamped UI will be implemented for Beta5 and all of it should be in place by Beta6.


Core projects

With regard to analytical tools and visualisations, the Portal is primarily an integration point for components developed by other institutions, particularly the Universities of Queensland and Newcastle.  In addition, some other projects form part of these core activities and the full listing appears below:

TitlePartner Institution
Statistical and Analytical tools and Visualisations for Urban Research DataThe University of Queensland
The Online 'What If?' Planning Support System ToolThe University of Melbourne
Metadata Management SystemThe University of Melbourne/Centre for Spatial Data Infrastructures &
Land Administration (CSDILA)
Scoping Study for Volumetric Modelling ToolsJ Barton Consulting
ENVISION Planning Support Tool
Swinburne University of Technology
ENVISION Scenario Planning Tool (ESP)CRC Spatial Information (CRCSI)
AURIN 3D Volumetric ModuleCSDILA
AURIN-Australian Stocks and Flows (ASFF) Data IntegrationMelbourne Sustainable Society Institute/University of Melbourne


Technical Architecture

The AURIN Technical Architecture follows a client-server, service-oriented (or resource-oriented) architecture model applied in a fashion that maximises re-use, scalability, and independence of its individual components. The aim is to establish a loosely coupled, flexible and extensible service-based architecture. Key to this architecture is that individual functional components can be reused in different situations. The implementation details of each component are hidden as much as possible from the external applications and end users. The design of each component allows for use of different programming languages and for their deployment on different operating systems. Key to this is the consistent specification and implementation of the component APIs (Application Programming Interface).

Individual components communicate through Web Service API calls, applying the Representational State Transfer (REST)-ful style of Web services. REST is a particularly useful style of encoding resources within a Web environment, where URLs (Uniform Resource Locators) act as the interface to resources (components). The syntax of the URLs is human readable and eases the complexity of development against REST services considerably.

The AURIN e-Infrastructure leverages JavaScript Object Notation (JSON – json.org) – a lightweight message format, for the encoding of the majority of its communication. JSON allows for hybrid messages with adaptive content. This is particularly advantageous for the complex data descriptions and formats to be passed around within the AURIN e-Infrastructure. The GeoJSON extension of JSON (www.geojson.org) in particular has been adopted for internal spatial data transfers. The overview of the internal architecture of AURIN is shown in the figure below.


AURIN Technical Architecture

AURIN Technical Architecture