Cloud only backup services though are typically solely dependent on the cloud for backups to occur. If you are not connected to the internet for some reason backups do not occur. Another challenge is getting that initial backup completed. Data in most cases has to transfer across the internet connection. Once that initial seeding is done most systems use either deduplication or delta differencing to send only changed segments of files across the internet. Which for the most part work fine. Another challenge is if between backup times you load a large set of new data which will lead to another long backup time. Cloud only services then are best suited for systems that are frequently connected to the internet and where there is only an incremental amount of daily change.
From a recovery perspective, as with backups, you typically have to be connected to the internet to be able to perform recoveries. Recovery of single files is not usually a problem in this type of configuration, even when using a 3G modem, most files can be restored in a few minutes. Recovery of an entire system, especially server class systems, can be problematic. There is simply too much data and deduplication help here. Some services have worked around this problem by offering a service of either shipping you the drive with the data on it or even a pre-built server. Even more interesting are services that actually store an image of your system in a virtual desktop environment and allow you to remotely access that virtual version if your system fails.
The challenges created by cloud only models though have lead to the creation of two other models; hybrid cloud backup and cloud enabled enterprise backup which we will cover in our next entry.
Track us on Twitter: http://twitter.com/storageswiss
Subscribe to our RSS feed.
George Crump is lead analyst of Storage Switzerland, an IT analyst firm focused on the storage and virtualization segments. Find Storage Switzerland's disclosure statement here.