The cloud is not ready for your data in the way you probably have come to expect from enterprise storage systems. But it is ready in other ways. The key to leveraging the cloud is to use it for what it’s good at doing, then wrap services and appliances around it to do what the cloud is not good at doing.
The cloud is ready to store your data reliably and make sure it would survive a disaster that would otherwise destroy your data. It would actually be quite expensive to bring that level of resiliency to the typical data center. Imagine the cost of a storage system that automatically and immediately copies your data to three different geographic regions. Add to that the cost and complexity of assuring constant data integrity.
Every cloud system has data integrity built into it, so you can be assured that an object that you stored in it today will be the same object months or years from now, despite anything that might try to change that.
Where cloud storage underachieves is performance, compatibility with existing systems, and typical data protection mechanisms. It may be resilient and have very strong integrity, but cloud storage systems are not designed for performance. It’s simply not possible, given that the closest copy of an object stored in the cloud is on the other side of an Internet connection. As to compatibility, everyone knows cloud storage systems are designed around object storage. One day applications will know how to write to something other than a POSIX-compliant file system, but that day is not here yet. Despite the incredible advantages that object storage brings to the table, this one fact probably holds them back more than any other.
Finally, there’s the all important topic of data protection. While the cloud will make sure your objects survive a tornado, and they will not be corrupted due to bit rot or hackers, they have nothing built into them to handle the number one reason we have backup systems – people. People accidentally delete files. People accidentally corrupt files by “fat fingering” something. Next thing they know they’ve accidentally overwritten the latest version of a file. Cloud storage systems aren’t yet ready to protect your data from that kind of damage, even though it’s the most popular reason you’ll need to restore something from backup.
This is why organizations wanting to leverage the cloud for storage must wrap other things around it to make it compatible with their workflows. Products that can make cloud storage look and feel like local storage go a long way towards making it useful. Just as important, they also provide the level of data protection that today’s environments are used to, so they can feel safe storing their files in the cloud.
My colleague George Crump discussed these issues in a webinar last week, and it’s now available on demand. You should check it out!