Petr Zapletal: From Reactive to Proactive: Predictive Scaling in Distributed Systems.
Updated: Jun 30

Petr is a Principal Engineer at Disney Streaming who specializes in the design and implementation of highly scalable, reactive and resilient distributed systems. He is a functional programming and open source evangelist and has expertise in the area of big data and machine classification techniques. Petr participates in the whole software delivery lifecycle; from requirement analysis & design, through to maintaining systems in production. During his career, Petr has worked for various companies from start-ups to large international corporations.
Petr's current interests are Reactive Systems, Distributed Streaming, and Deep Learning. Petr is also an author of #ThisWeekInScala and a seasoned conference speaker.
From Reactive to Proactive: Predictive Scaling in Distributed Systems.
As distributed systems continue to evolve and grow in complexity, effective resource management becomes a critical challenge. Scaling these systems to meet varying workloads while minimizing resource waste is crucial for achieving optimal performance, reliability, and cost-effectiveness. Traditional autoscaling approaches often rely on reactive techniques, resulting in delayed responses and suboptimal results.
This talk explores the concepts of predictive scaling, where artificial intelligence techniques are leveraged to proactively anticipate workload patterns and dynamically adjust system resources. We will discuss the underlying principles, architectural and design considerations, and also practical implementation guidelines and recommendations.