This week’s news of the T-Mobile Sidekick Disaster is once again raising concerns about the viability of cloud computing and the need to have a strategy for data replication and archival. One article I saw went so far as to suggest that the Sidekick debacle was all part of an evil Micrososft plot to discredit “the big switch” to the cloud. While I hardly see this as a “sky is falling” moment, it does put additional pressure on SaaS/Cloud application and platform vendors and customers to be more transparent about data security, privacy, and all around safety.
I received this email today from a Salesforce customer that summarizes some of what you might want to consider when it comes to SaaS /Cloud data replication:
“This backup solution must be able to:
- Backup all data from Salesforce from all standard objects, custom objects, and their related lists.
- Restore data to Salesforce if/as/when necessary.
It would be important that we can restore without creating duplicate records and that we have a choice whether to keep or purge records which are newer than the backup.
The ideal solution would also:
- Periodically perform the backup operation entirely automatically at a specified time or interval.
- Allow us to restore to a specific date and time (instead of to whatever data has most recently been replicated.)
Obviously, all the data which is being backed up should be saved within a secure environment and there should be an administrative interface which will enable administrators to:
- Determine when backups should happen.
- Verify that backups have happened.
- Manually start a backup.
- Restore from backup.”
How important is data replication to your SaaS/Cloud strategy? Is it all about having a cloud backup or is the use case about reporting, compliance, governance, or something else?