What is the difference between S3 Standard-IA and S3 One Zone-IA?
Both are for data accessed rarely but needed quickly. Standard-IA stores copies across at least three Availability Zones. One Zone-IA keeps data in a single AZ at lower cost, so it is lost if that AZ is destroyed.
Explain it simply
Standard-IA is keeping your old photos in a bank locker and also a second copy at your uncle's house in another city. One Zone-IA is keeping them in one locker only. Cheaper, but if that bank has a fire, they are gone.
Example
Standard-IA One Zone-IA
AZs used : 3 or more 1
Resilience to AZ loss: yes no
Cost : lower than Standard about 20% lower than Standard-IA
Retrieval fee : per GB retrieved per GB retrieved
Minimum duration : 30 days 30 days
Minimum object : 128 KB billed 128 KB billed
Good for : backups, older data re-creatable data, secondary copiesBoth give millisecond access, unlike Glacier classes. Both charge a retrieval fee and a minimum storage period, so they do not suit data you read often or delete within days.
A sensible use of One Zone-IA is a thumbnail or transcoded copy that can be rebuilt from the original, or a second copy of data that already exists elsewhere.
What to say in the interview
"Standard-IA spreads data across multiple AZs and One Zone-IA uses just one for a lower price. I use One Zone-IA only for data I can re-create or already replicate, and Standard-IA for important data that is read rarely."
Common mistakes
- Putting the only copy of important data in One Zone-IA.