A friend of mine is the CFO for a business that provides online services for media content providers. He once told me their internal IT motto regarding storage is, “if it spins, fill it.” This, sadly, is the case for many businesses, especially if they have a heavy emphasis on file-level, aka “unstructured” data. Some industry analysts predict storage demand for unstructured data will increase by more than 500% in the next five years. If this prediction is true, every 1 TB of unstructured data you have now will be 6 TB in five years.
There are two traditional ways of dealing with the problem of uncontrolled data growth:
1. Gaze into a crystal ball and try to forecast storage demand into the future, usually for three to five years. Then budget for, shop for, acquire and install enough up-front capacity to handle the projected future demand, and hope like hell you are right. This approach has several significant drawbacks:
- Paying for tomorrow’s capacity at today’s prices. Storage prices, in terms of cost-per-GB, have been on a steady or even accelerating decline for decades. Why pay $X per empty raw GB now when the cost might be 30% or even 50% less in three years?
- The business opportunity cost of obligating capex funds for big up-front storage acquisitions, when those funds may be better spent elsewhere.
- If you care about these things, you are sending carbon dioxide up a smokestack every second you have those empty platters spinning. If you don’t care about that, you should at least care about the opex cost of the power & space those empty platters are using.
2. If forecasting long-term needs isn’t your thing, then you are faced with a never-ending treadmill of frequent storage expansion, which at the very least is potentially disruptive to IT operations. At the worst it can expose you to data loss or operational failures, depending on whose storage architecture you are using.
Now there is a better way to handle this problem. Online storage provider Nasuni offers a pay-as-you-go model whereby encrypted data is automatically replicated & stored at a highly secure offsite location. This is commonly called “cloud storage”, a service available from various providers, not just Nasuni. The Nasuni offering, however, has numerous interesting wrinkles that make it a compelling solution:
- Nasuni uses a local filer appliance as a cache, so that there is little or no noticeable performance degradation. And the appliance’s CIFS/NFS interface makes it very easy to install and use on a network.
- Nasuni filers can share encryption keys, enabling a “pooled” storage configuration whereby data can be easily seen and shared among multiple physical locations. This multi-site access is controlled via a central admin console.
- Data replicated via the Nasuni filer is securely and redundantly stored in the cloud. This means that data written to the filer no longer requires local backup, which can streamline backup operations for all other data while freeing up space for data not suitable for cloud storage.
- Capacity is provisioned in 1 TB increments that can be expanded on the fly with a simple phone call. There is no theoretical limit to the online capacity provisioned behind the filer(s).
- Nasuni transparently manages the relationship with the cloud storage service (Amazon S3) so that there is only one bill to pay.
- Nasuni has a standard 100% uptime SLA.
So there is, finally, an elegant, economical solution to the “if it spins, fill it” problem of unstructured data growth. If you are struggling with this problem, you would do well to give the Nasuni solution a good hard look. Contact us for a demo or a free evaluation of the Nasuni solution.