Community Sessions

The Rapid Analytics for Disaster Response project (RADR), funded under USA's Department of Energy's Cybersecurity, Energy Security, and Emergency Response (CESER) program, collects satellite imagery and provides damage assessments to first responders during natural disaster events. Catch this talk to hear about our journey into the cloud; starting with traditional VM server infrastructure with manual deployments, and learn about our evolution over the years to 100% serverless architecture and Architecture as Code for automated deployments!

AWS recently released AppSync Events, a standalone, managed WebSocket feature. But is that really what we've been looking for? WebSockets, even managed ones, are hard! In this session, get a glimpse of how the service works, see what's still on developers to take care of, and look at some performance benchmarking.

People think that Serverless is just for simple web apps and not for "real software". In this talk I'll show you how we built a serverless platform that deploys, manages and controls traditional video editing applications.

Traditional IT operations is overwhelmingly reactive - humans respond to incidents and tickets. There is so much noise! Alerts, tickets, audits, chat threads. What if ops could catch the small things before they turn into big problems? What if we could use events and serverless to remediate issues?

Vadym Kazulkin explores the critical scalability considerations and quotas for AWS Serverless services, dispelling the myth of unlimited scaling. He covers key services like Lambda, DynamoDB, Aurora Serverless, API Gateway, and S3, highlighting quotas and concepts such as concurrency, throughput, and retry strategies. Vadym also discusses how to design architectures that account for these limitations using distributed systems principles like the token bucket algorithm. Finally, he examines decision-making for Serverless applications, including database and messaging service choices.

DynamoDB makes it super fast to get started. But be careful getting started too quickly! Even with up front schema design and knowing your access patterns, things can change as your application grows. In this session, learn some hard and fast rules about getting the most out of DynamoDB, from how you structure your data to how you tune its dimensions.

What happens when your personal project runs out of space on that piece of decommissioned hardware you scraped your pennies together to buy? The cloud represents an opportunity to leverage elastic compute at a low cost to get your idea up and running with a minimum of time and cost. Follow along as Greg discusses the optimizations he made to make a part of his passion project that outgrew his home hardware run in the cloud, processing millions of records for less than $4 a day!

Eventbridge can be a key part of an event driven system, but how do you successfully govern this? We’ll look at ways to visualise and govern your event driven systems, and how schemas can help speed up development. We’ll take a look at some examples of using Schema Registries to easily share schemas across an organisation and how code generation can get you quickly consuming your events.

This live coding session shows you how simple it is to build an app with arc.codes. Get your creativity flowing and toss out features during the session as Brian builds them before your very eyes!

At the end of 2024, we all have our favorite livestreamers. Watching somebody stream is as common as watching our favorite tv shows. So you should do it too! But that might be easier said than done. In this talk, we'll cover what goes into an engaging livestream, talk about the special equipment you need (or lack thereof), and how to bring the most to your viewers.