Designing High-Throughput Event Sourcing Systems with PostgreSQL
Designed for Senior Backend Engineers and System Architects leading event-driven microservices initiatives in rapidly scaling fintech or SaaS environments to spark real collaboration and high-energy learning.
A 90-minute hybrid workshop with hands-on labs. Most attendees have deep RDBMS and distributed systems backgrounds but are under pressure to deliver resilient, scalable event-driven architectures without easy access to Kafka or kinesis. Many have experienced reliability or performance issues with ad-hoc event sourcing built atop PostgreSQL.
Event-Driven Detective: Log Sleuth
Kick off with a fast-paced mystery: Attendees receive a real-world production event log snippet (e.g., JSON events with out-of-order writes and duplicates). Their challenge is to spot the hidden bug that broke downstream processes. This primes their curiosity about event sourcing edge cases.
Tap to view the full activity.
Why this works
Starting with a puzzle boosts intrinsic motivation, activating prior knowledge and focusing attention on gaps they didn’t know they had.
Mythbusting: PostgreSQL Under Pressure
Run a live poll presenting three commonly held beliefs about PostgreSQL event sourcing (e.g., 'PostgreSQL replication guarantees no data loss under crash'). After voting, debunk each with concise evidence and real post-mortem anecdotes.
Tap to view the full activity.
Why this works
Surfaces preconceptions, then creates a teachable moment by helping participants update their mental models using concrete evidence.
Safe Sandbox: Schema Sketch
Guide participants through a low-stakes, hand-drawn event schema design—no IDEs, just markers/whiteboard or online sketch tools. Invite folks to annotate how they’d store, partition, and replay events for a simple banking transfer scenario.
Tap to view the full activity.
Why this works
Reduces performance anxiety, makes design approachable, and encourages creativity before syntax and tooling enter the conversation.
Throughput Throwdown: WAL Race
Facilitate a simulated write-velocity contest: teams use a visualizer (or facilitator-provided script output) showing WAL (Write-Ahead Log) generation rates as different event ingestion strategies are simulated. They bet on which strategy will scale longest before bottlenecking.
Tap to view the full activity.
Why this works
Competition energizes and cements the reality of bottlenecks—participants see tradeoffs, not just hear about them.
‘Oh No!’ Outage: Recovery Dilemma
Present a real-world dilemma: ‘PostgreSQL node loses all recent WAL after a burst of event writes during a flash sale. What data can you recover—and what’s forever gone?’ Assign roles (Ops, Dev, Product Owner) and spark a debate on priorities and tradeoffs.
Tap to view the full activity.
Why this works
Fosters systems thinking and empathy for cross-functional constraints, making abstract tradeoffs tangible under pressure.
PostgreSQL & Me: Reflection Wall
Invite each participant to post a sticky (physical or digital) answering: ‘What’s one event sourcing pain I’ve hit with PostgreSQL, and what change will I try next sprint?’ Close with a brief shareout and actionable commitment.
Tap to view the full activity.
Why this works
Active reflection anchors learning in personal experience and motivates real change by connecting insights to next actions.
Sign up to unlock 3 more activities
Get the full pack, facilitation flow, and more ready-to-run ideas.