Do you use dead letter queues, EventBridge archives or pub/sub replays as part of your application architecture? What if streams like Kafka or Kinesis were part of the operational stack? Some might also suggest provisioning additional topics or streams for reliable processing; however, I would argue that this is an anti-pattern that attempts to replicate a queuing strategy. While aspects of reliability and consistency are certainly addressed, they often come at the cost of maintainability.

Fundamentally, eventually consistent paradigms address only part of the problem, primarily at the storage layer; it still requires a lot more to solve consistency at a broader business application layer. In principle, the key levers associated with software businesses are business objectives, operations, and maintainability. While the bar for business objectives can never be lowered, operations and cost to maintain are always inversely proportional to each other and will be the major challenge for most organisations.

Join me as I introduce you to durable execution, where neither lever is compromised, nor is any ocean boiled.

Technical requirements: general software system understanding, distributed systems, event driven architectures

Target audience: CTO's, developers, product people, software philosophy and enterprise scale development

preferred session duration - 45 mins