-4

I came across this question and thought it is simple, my choice is A but it turns out the right answer is C.

Can anyone share me with any explain please? Thank you very much.

Which of the following approaches provides the lowest cost for Amazon Elastic Block Store snapshots while giving you the ability to fully restore data?

A. Maintain two snapshots: the original snapshot and the latest incremental snapshot
B. Maintain a volume snapshot; subsequent snapshots will overwrite one another
C. Maintain a single snapshot the latest snapshot is both Incremental and complete
D. Maintain the most current snapshot, archive the original and incremental to Amazon Glacier.
mdivk
  • 3,545
  • 8
  • 53
  • 91
  • `If you make periodic snapshots of a volume, the snapshots are incremental. This means that only the blocks on the device that have changed after your last snapshot are saved in the new snapshot. Even though snapshots are saved incrementally, the snapshot deletion process is designed so that you need to retain only the most recent snapshot in order to restore the volume. ` https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ebs-deleting-snapshot.html – Matus Dubrava Dec 20 '19 at 21:38

1 Answers1

2

Think of an Amazon EBS Snapshot as an index to a backup of the blocks on a volume.

If a Volume has blocks 1, 2 and 3 used, then taking a snapshot will copy those three blocks to S3 and the snapshot will be an index to those three blocks.

Then, if the contents of Block 1 is changed and a Snapshot is taken, then only the updated block 1 will be copied to S3. The second Snapshot will point to the new Block 1 and the old Blocks 2 & 3. The old Block 1 will still be maintained in S3 because the first Snapshot is still using it. If the first Snapshot is deleted, then the old Block 1 will be discarded.

That is why the lowest-cost option is to maintain only the latest Snapshot, since it contains the smallest about of backed-up blocks.

Snapshots are incremental in that they backup whatever has not been previously backed-up by another (existing) snapshot.

Update: See: AWS re:Invent video that explains the 'incremental' concept of Amazon EBS volumes

John Rotenstein
  • 241,921
  • 22
  • 380
  • 470
  • Thanks John, as you indicated, there needs to be a very initial snapshot to serve as the base snapshot, and then you need to maintain only the latest snapshot, that's what I thought: base + incremental, two snapshots. What am I missing here? – mdivk Dec 20 '19 at 23:16
  • Every snapshot, including the first one, is incremental. There is no "base" snapshot. Again, don't think of a Snapshot as the back-up — think of it as an _index_ to blocks that have been backed-up. The first snapshot can be deleted and any blocks used by other snapshots will remain in S3 as backed-up blocks. Only blocks referenced in _any_ snapshot will remain in S3. (Hurts the head, eh?) – John Rotenstein Dec 20 '19 at 23:28
  • Thanks John, as per you said, then the S3 bucket for backup of the snapshot, though only the latest snapshop is needed, but if the old snapshots are not getting deleted, then the S3 must keep a copy of the block that is used by the old snapshots. Am I understanding correctly now? If yes, wouldn't that mean S3 bucket would get bigger and bigger? What is the best practice? – mdivk Dec 21 '19 at 00:12
  • Your understanding is correct. The more snapshots, the more storage space is required because there is data that is different between each snapshot. The above question is asking for the lowest-cost approach while being able to fully-restore data. The answer is to keep only the latest snapshot since a single snapshot would be storing less data and can also be used to fully restore the volume. – John Rotenstein Dec 21 '19 at 03:25
  • alright, thank you for your enlightening, it is very much appreciated here. – mdivk Dec 21 '19 at 13:08
  • I've added a link to a re:invent presentation that demonstrates the concept. – John Rotenstein Dec 28 '19 at 02:19