Cloud Data Protection FAQ
- 26 Sep 2022
- DarkLight
Cloud Data Protection FAQ
- Updated on 26 Sep 2022
- DarkLight
Article summary
Did you find this summary helpful?
Thank you for your feedback
Frequently Asked Questions
What is a Cohesity storage domain?
A storage domain is a logical object that contains all of the data written from backups. This includes the physical data from the backups themselves, but also overhead such as data resiliency, an index of backup jobs, and garbage collection.
What is a protection job?
A protection job is a logical object that defines what needs to be backed up, how it should be backed up, and what retention policy should be used to maintain those protection jobs.
What is a protection job snapshot?
A protection job snapshot is a synonymous term for a backup job.
How do I read the backup reporting for storage domains?
| Logical Data | Also known as front-end data or hydrated data. This number represents the total amount of data that was backed up if compression and deduplication did not take place. |
| Data In | This is the estimated amount of data that was sent from a backup source to Cohesity before deduplication and compression. |
| Data Written | This is the true utilization after taking deduplication and compression into account. |
| Resiliency Impact | This is the utilization overhead that is intended to protect Data Written in the event of a hardware failure. The normal expectation for resiliency footprint is 50% of data written. |
| Storage Consumed Local and Cloud | This is the total utilization when adding Data Written and Resiliency Impact. This is the utilization that is reflected as the Storage Domain Quota. |
As seen from the illustration, this example shows that the storage domain (Storage Consumed Local and Cloud) is utilizing 151.9GiB of space. Out of that 151.9GiB, 100.7GiB is Data Written. Data Written is the actual footprint from your backups. The erasure coding overhead is represented by Resiliency Impact and shows to be about 50% of the Data Written number as expected.
Why is Resiliency Impact 50% of Data Written?
This is the expected overhead, as Expedient standardizes a 4+2 Erasure Coding scheme.
Are there any other reporting metrics that provide additional granularity?
Unfortunately, there is no reporting that is available that provides granularity beyond the scope of the above report. This is due to deduplication,as all protected workloads are part of the same deduplication domain and may share the same blocks between objects.
I manually deleted some backup snapshots and have not seen Storage Domain utilization drop. Why?
Space reclamation is not immediate despite the snapshot being deleted prior to a retention date. This is to be expected, as Cohesity runs a series of complex internal operations to validate and earmark blobs of data before deletion. This internal process can take 5-7 days before space can be large reclaimed.
Was this article helpful?