Cancer Council NSW

Migrating Cancer Council to Azure Cloud

Executive Summary

Australia’s Daffodil Centre, a joint venture between the non-profit Cancer Council NSW and the University of Sydney. While Cancer Council had adopted Microsoft 365, Dynamics 365, and Azure for the rest of the non-profit’s operations, the Daffodil Centre modelling team were still using on-premises, legacy infrastructure to run complex models. Migrating to Azure has improved model simulation time from a week to half a day.

The Client

Australia’s the Daffodil Centre, a joint venture between the non-profit Cancer Council NSW and the University of Sydney, is focused on finding the most effective ways to reduce cancer death and illness. The non-profit is transitioning its cancer modelling research efforts to the Azure cloud to accelerate the pace of research into prevention, early detection, treatment, and care. Increasing the speed and scope of research-based insights could help the organisation save even more lives.

The Problem

While Cancer Council had adopted Microsoft 365, Dynamics 365, and Azure for the rest of the non-profit’s operations, the Daffodil Centre modelling team were still using on-premises, legacy infrastructure to run complex models.

The research teams had to schedule who was going to use the organisation’s 48 virtual machines and sometimes waited weeks for their turn. If time-sensitive questions or issues cropped up, turnaround timing lengthened, and scheduling congestion increased.

The on-premises infrastructure used ageing hardware and software which required frequent repairs and patches to maintain security. The non-profit faced a choice: Either heavily invest to upgrade the ageing infrastructure or transition to the cloud.

The Solution

An organisation-wide cloud-first strategy led the Daffodil Centre modelling researchers to work with cubesys to test the cloud’s ability to handle complex scenario-based modelling studies. The team successfully created a use case based on a prostate cancer model in Azure Batch and automation delivered by Azure DevOps.

Cloud adoption gives the research teams access to the tools that will accelerate their work and take away the constraints that slow them down with their current on prem systems.

With the limitations of legacy technology, research teams had to be extremely careful in what details or new scenarios they added to their research. Removing that constraint empowers the team to provide nuanced and timely information to its partners.

The Outcome

In the old on-premise environment, It could take about a week to run a model simulation, but by shifting the work to the cloud’s unlimited compute power, researches can now expect to cut that run time to a half-day. They will be able to run as many of these models in parallel as they like, instead of having to queue up one after another, giving research teams more flexibility and control over work timelines.

By running fully automated models on demand, researchers can reduce their administrative overhead and spend more time on research tasks.  IT administration and system maintenance has been removed or vastly reduced has as the associated cost.

The cubesys team is now in the process of recreating their two most complex models, for bowel and cervical cancers, in the Microsoft cloud.

Software and Services

• Microsoft Azure
• Azure Batch
• DevOps Automation