Digital Institute

e-Science Central

e-Science Central

e-Science Central allows for the development of portable analytics applications that can be deployed on dedicated hardware or in a Cloud-based environment.

It is able to make use of a variety of scalable storage drivers (AWS S3, Azure Blobstore, Filesystem) and automatically versions all data. Security is paramount and data can be encrypted both at rest an in transit. Within e-Science Central users must be given explicit permission to be able to view documents and run applications.

The differentiating factor for e-Science Central is the ability to analyse data in addition to storing it. Users can define workflows that can use both common workflow blocks provided in the installation and define their own custom blocks if they require something more specialised. There are over 150 common workflow blocks which perform tasks such as

  • data manipulation
  • statistics
  • visualisation
  • machine learning

Workflow blocks can be written in Java, R, Javascript and Octave. They can also make use of third party applications and libraries where required.

The scalability of the system is provided by the Workflow Engine which enacts each workflow. This engine manages the installation of dependencies, transfer of data and execution management. In the VENUS-C project we had 200 Workflow Engines running concurrently to build models of chemical compounds demonstrating the scalability of the platform.

Many projects use the e-Science Central API to access e-Science Central programatically. For instance, Stroke Rehabilitation through Games developed a number of websites that used the e-Science Central API for data storage, processing and user management. The Low Impact Light Rail project used a desktop application to perform similar activities from the API.