What is Consistent Hashing? How is it used in Distributed Systems?
Consistent Hashing
Consistent Hashing is a partitioning technique that maps both keys and nodes onto a logical hash ring. A key is assigned to the next node on the ring according to the chosen traversal rule. Its main advantage is that when nodes are added or removed, only a relatively small portion of keys need to move compared with naive hashing based directly on modulo node count.
- 1
Reduces data movement when cluster membership changes.
- 2
Commonly used in distributed caches and partitioned storage systems.
- 3
Virtual nodes improve distribution and reduce hot spots.
- 4
Replication can assign a key to multiple successive nodes.
- 5
The hash function and node placement influence load distribution.
- 6
Consistent hashing does not automatically provide replication, consistency, or fault tolerance; those are separate system-design concerns.
Scenario Questions
0-2 years experience
- 1We have a small cache cluster of four nodes using consistent hashing. If we add a fifth node, how does the key distribution change?
- 2Given a simple hash ring with eight slots, where would the key 'user123' be placed? Walk through the steps.
2-5 years experience
- 1Our service uses consistent hashing for sharding data, but after a node failure we see a spike in latency for certain keys. What could be causing that and how would you investigate?
- 2We need to decide between using virtual nodes versus plain nodes in our consistent hash ring. What are the trade‑offs and which would you pick for a system with 1,000 keys and frequent node churn?
5-8 years experience
- 1Design a distributed logging system that writes logs to storage nodes using consistent hashing. How would you handle node addition, removal, and rebalancing while keeping write latency low?
- 2Explain how you would migrate an existing key‑value store that currently uses modulo hashing to consistent hashing without downtime.
8+ years experience
- 1Our company is migrating from a monolithic cache to a multi‑region, globally distributed cache that uses consistent hashing. What architectural considerations, data‑migration strategies, and operational safeguards would you put in place?
- 2Discuss the long‑term maintenance challenges of using consistent hashing across multiple data centers, especially regarding clock skew, network partitions, and hot‑spot mitigation.
Follow-up Questions
- What steps would you take to rebalance the ring after adding a node?
- How does using virtual nodes affect key distribution variance?
- Can you describe how you’d detect and handle a hot spot on the ring?