Key Stats
- Increased modelling ignition resolution and number of weather scenarios by a factor of ~70x compared to previous implementation"
- 100% – Automation of data pipelines and post-processing workflows, removing human intervention and increasing performance.
- Key Technologies – AWS (Amazon SageMaker, Lambda, Athena), Python, Phoenix RapidFire.
- Core Principles – Scalability, Automation, Cloud-Native Design, and Cost-Effectiveness.
Overcoming Critical Technical Barriers: DEECA's existing bushfire risk framework was hampered by outdated data and processing bottlenecks. Kablamo engineered a solution to eliminate these constraints, enabling a new level of accuracy and scale.
Advanced Cloud and ML Integration: We developed CloudFARM,' a bespoke AWS platform that hosts the Phoenix RapidFire model in a scalable cloud environment and enhances its predictions with the power of Amazon SageMaker for advanced machine learning.
Delivering a Future-Proofed, Scalable Solution: The resulting platform is not just a modernisation but a long-term asset. Its serverless architecture and automated workflows provide DEECA with a powerful, cost-effective, and scalable tool for protecting Victorian communities.
The Approach
Kablamo was engaged to design and deploy a cloud-native, fully scalable AWS platform to underpin the next generation of DEECA's risk modelling. Nicknamed 'Cloud Farm,' the solution was architected with several key principles in mind:
- Cloud-Native Modernisation: We migrated the Phoenix RapidFire model, originally designed for desktop use, to a high-performance AWS cloud environment.
- Machine Learning Integration: The platform was designed to incorporate advanced machine learning capabilities using Amazon SageMaker, allowing for more sophisticated analysis and enhanced predictive accuracy.
- Automation and Scalability: We developed automated, serverless data pipelines using services like Amazon Lambda and Amazon Athena. This approach was designed to automatically scale to accommodate massive increases in data and remove the post-processing bottlenecks of the previous system.
- Cost-Effectiveness & Visibility: The solution was built to be as efficient as possible, automatically spinning down unused resources and optimising storage. A rich web interface was included to give operators full visibility and control over jobs, inputs, and outputs.
The Results
Leveraging our deep expertise in cloud engineering and data science, Kablamo delivered a robust, scalable, and cost-effective platform that has fundamentally transformed DEECA's capabilities. The CloudFARM solution allows scientists and fire chiefs to swiftly create, simulate, and repeat complex scenarios with high-resolution data.
The fully automated, AWS cloud-native architecture has eliminated previous scaling constraints, offering near-limitless capacity in a cost-effective manner. This ensures that DEECA's critical risk metric can now be produced using the most accurate and powerful digital technologies available, future-proofing a vital public safety function for years to come. Kablamo continues to provide operational support and ongoing product development through our Product Care services.
