Reducing the environmental footprint in nautical transport with F# & Serverless
summary
A cloud-backed Weather Service aggregates Copernicus, DMI, FMI and MET providers to feed a vessel-side Fuel Pilot used across 37 vessels and ~25,000 sailings, claiming tens of percent CO2 savings. Implementation uses F# on .NET 6, Azure Functions (serverless) for event-driven jobs, App Service containers for Java conversion, Cosmos DB with immutable records + TTL, and an SDSlight netCDF adapter. Key takeaways: interpolate missing wind/current, send cache diffs over satellite to save payload, avoid serializing discriminated unions, and throttle functions to control RU costs.
AI-written summary · about
chapters (16)
- 0:00Introduction and context
- 1:32Problem and initial approach
- 4:48Stena Weather Service design
- 7:22Technical implementation
- 10:35Technology stack choices
- 13:03Domain modeling with F#
- 16:26Handling netCDF files
- 20:27Data reading implementation
- 24:19Storage strategy
- 28:11Serialization lessons
- 31:23Background jobs and Azure
- 35:28Optimizing Azure Functions
- 40:00Scaling and performance
- 46:41Error handling strategy
- 49:21Monitoring and infrastructure
- 51:47UI and project conclusion
related talks
01h 17mReducing the Environmental Footprint in Nautical Transport With F# & Serverless
11mTotal Rewrite: A Story About F# and Azure Functions in Production
44mMicroservices Without Servers
17mJavaScript Saves the World
22mUnFaaSener: Latency and Cost Aware Offloading of Functions from Serverless Platforms
39mServerless at the end of the Universe
59mAzure Serverless for JavaScript TypeScript Folks
30mEvolving Serverless Architectures