Cloud Edition level of data redundancy

[Machiel van Hooren]

I'd like to know at what level of redundancy our data is stored in the cloud.

I read somewhere that Cloud Edition relies on Azure to provide data redundancy, however, Azure offers various levels of redundancy. 

See https://docs.microsoft.com/en-us/azure/storage/common/storage-redundancy 

So is our data locally-redundant, zone-redundant or even geo-redundant?

[calbzam]

Hi,

Could you open a ticket in support@plasticscm.com for furhter details?

Regards,

Carlos.

[Edwige]

Hi! I would also be interested in this information. Also, in case a repository was deleted (by mistake or by malicious intent), how easy would it be to recover it?

[calbzam]

Hi,

We actually store our repository data on Google Cloud Platform. 

We perform nightly backups of the repository metadata and retain those for at least 7 days. 

The repository data itself is stored in Binary Large Object (BLOB) files on Google Cloud Storage persistent disks. According to Google's own documentation:

Objects written to Cloud Storage must be redundantly stored in at least two different availability zones before the write is acknowledged as successful. Checksums are stored and regularly revalidated to proactively verify that the data integrity of all data at rest as well as to detect corruption of data in transit. If required, corrections are automatically made using redundant data.

The integrity and the error recovery from those copies are handled by Google automatically.

In the event of a hardware or system failure in a GCP data center, there could be some downtime as the SLA of GCP that for our configuration is >= 99.9%. However, the integrity of the data, as noted above, is protected. 99.9% uptime translates to a maximum total downtime of fewer than 9 hours per year. 

Regards,

Carlos.