Lessons from the Field: Scale, Availability, and Performance
Join Aerospike AWS, Adjust and Miro for an evening of technical talks focused on the challenges of running real-time, business-critical systems at scale.
Hear firsthand lessons from engineers managing large datasets, database migrations, and cloud infrastructure availability in production. From Miro's Redis-to-Aerospike migration on Kubernetes to AWS best practices for capacity planning and resilience, speakers will share practical insights gained from operating demanding workloads where performance and reliability matter.
Connect with fellow engineers, architects, and technology leaders from Berlin's tech community over drinks and networking.
Agenda
17:30 – 18:00 - Welcome & networking
18:00-18:10 - Speed vs Capacity: Designing Data Systems for Real-Time Scale
Dirk Möller, Regional Lead Germany and EE, Aerospike
In high-performance data systems, balancing RAM and disk is key to real-time performance at scale. This session explores the technical nuances of database architecture in today’s AI-driven era, where workloads demand low latency, massive scale, and adaptive infrastructure. Drawing on real-world, mission-critical systems, we’ll show how to achieve the right mix of speed, scale, cost-efficiency, and reliability - designing architectures that meet today’s real-time demands while staying sustainable and affordable.
18:10 – 18:30 - Scaling Beyond Millions of Transactions
Leonardo Muñoz, Platform Engineer, Adjust
A behind-the-scenes look at how Adjust powers high-throughput, low-latency workloads at scale using a distributed database platform running on bare metal infrastructure. Leonardo will explore the architectural and operational decisions behind the deployment, the challenges of managing dedicated hardware at scale, and the lessons learned from years of production experience supporting mission-critical workloads.
18:30 – 18:50 - Pod Ready ≠ Data Safe: Migrating a Critical Dataset on Kubernetes
Yuriy Chernikov, Staff Software Engineer, Miro
Miro migrated one of its most critical datasets from a legacy caching platform to a modern distributed database running on Kubernetes - and not everything went according to plan. This practical talk covers what the migration actually required, the design decisions that made failures survivable, and the operational lessons that only emerge when managing datasets measured in terabytes.
18:50-19:10 - Managing Amazon EC2 Capacity and Availability
Nati Cohen, Sr. Specialist Solutions Architect, Compute, AWS
Modern data-intensive applications require infrastructure that delivers high performance, scalability, and resilience. Organizations often seek to leverage the latest storage-optimized EC2 instances, select larger instance sizes for improved performance, place infrastructure close to data producers and consumers, and scale efficiently as business demands grow.
In this session, we'll explore how to balance these performance requirements with cost optimization and capacity availability. We'll review the various EC2 capacity and reservation models, including Capacity Reservations, On-Demand Instances, Savings Plans, and more. Learn how to combine these options to optimize cost, performance, and availability for large-scale production workloads
19:00-20:00- Networking