Consistency in Distributed Systems
What you'll learn: You'll understand what strong consistency means in distributed systems and why achieving it is challenging but valuable.
The Core Idea
In a distributed system with multiple nodes (servers, databases, etc.), strong consistency guarantees that all nodes see the same data at the same time after an update. When a write completes successfully, every subsequent read from any node returns that updated value—no exceptions, no delays.
Think of it like a classroom announcement system: if a teacher updates tomorrow's exam time on the main board, strong consistency means every classroom's clock instantly reflects that change. No student sees the old time once the update happens.
Why This Matters
Without strong consistency, different nodes might temporarily (or permanently) show different values for the same data. Imagine checking your bank balance on your phone (50. Which is correct? Strong consistency eliminates this confusion.
The Trade-Off
Achieving strong consistency requires coordination between nodes. Before confirming a write, the system must ensure all relevant nodes agree on the new value. This coordination:
- Takes time (increased latency)
- Requires communication across the network
- May block operations if nodes can't reach each other
Remember network partitions and lack of global clock from previous lessons? These make strong consistency expensive or sometimes impossible to maintain during failures.
Strong Consistency in Practice
Strong consistency means:
- Linearizability: Operations appear to happen instantaneously at some point between their start and completion
- No stale reads: You never read outdated data after a successful write
- Sequential ordering: All nodes agree on the order of operations
Key Takeaway: Strong consistency ensures all nodes see identical data simultaneously after updates, providing predictability at the cost of coordination overhead and reduced availability during network issues.