What consistency guarantees does object storage give, and why does it matter?
What they are really testing: Whether you know strong vs eventual consistency for object stores, that some historically returned stale or 404 right after a write, and the bugs that causes.
A real interview question
What consistency guarantees does object storage give, and why does it matter?
What most people say
drag me
“Object storage is consistent, so a read always returns what you wrote.”
Not universally true, object stores have historically been eventually consistent, where a read right after a write could be stale or a 404. Assuming strong consistency everywhere causes intermittent bugs. You must check the guarantee.
The follow-ups they ask next
What bug does eventual consistency cause in a write-then-read workflow?
Right after writing an object, an immediate read or list may return stale data or a 404 because the write has not propagated. The workflow fails intermittently and is hard to reproduce since it depends on timing.
How do you handle an eventually-consistent store safely?
Do not rely on immediately reading/listing what you just wrote, add retries with backoff, or pass the data forward directly in the pipeline instead of round-tripping through the store and racing propagation.
What the interviewer is listening for
- Distinguishes strong vs eventual
- Knows the write-then-read 404 gotcha
- Confirms the guarantee + designs for eventual
What sinks the answer
- Assumes always strongly consistent
- Unaware of eventual-consistency bugs
- No mitigation for eventual
If you genuinely do not know
Say this instead of freezing. Reasoning out loud from what you do know beats silence every single time, and a good interviewer is listening for exactly that.
“It depends on the store: [strong read-after-write means a read sees the latest write immediately; eventual means a read/list right after a write may be stale or a 404]. The gotcha: [write-then-immediately-read pipelines fail intermittently under eventual consistency]. [Modern stores (S3 since 2020) are strong, but confirm, do not assume]. Under eventual, [do not rely on immediate read/list, use retries or pass data forward].”
Keep going with storage
Foundation
What is the difference between durability and availability for storage?
Foundation
What is the difference between backup and replication, and why is replication not a backup?
Junior
Explain the difference between object, block, and file storage, and give one good use case for each.
Junior
How does object storage achieve such high durability?
Junior
For block storage, what is the difference between IOPS and throughput, and how does it affect volume choice?
Mid
Design durable, highly available storage for user-uploaded images. Which service do you choose and why?
Knowing the answer is not the same as recalling it under pressure
Sign in to send the questions you fumble to spaced recall, so they come back right before you would forget them, and learn the concepts behind them with hands-on labs.
Start free