The major public cloud storage providers now all offer multiple tiers of storage. Like in the data center each successive tier becomes less expensive per GB. But unlike the data center it becomes more costly per access. IT should look for solutions that leverage each of the available tiers but also provides insight to ensure the right data is placed on the right tiers at the right time.
What are the Cloud Tiers?
Most cloud storage providers have at least three basic tiers. First is a production tier, used for cloud-based applications that need high-performance. These applications will also continuously access the data on this tier. Pricing for the production tier tends to be the most expensive per-GB, but the least expensive per access if there is a charge at all.
The second tier is essentially an active archive tier. Data stored on this tier supports immediate access but not necessarily high-performance or equivalent SLAs to the primary tier. Pricing on this tier is less expensive per GB than the high-performance tier, but it may have a more expensive cost-per-access.
The third tier is a deep archive. Users are rarely if ever expected, for the long term, to access data on this tier again. The cost per GB is the cheapest of the three cloud tiers, rivaling tape storage. However, the cost to access data from the cold archive tier is the highest of the three tiers.
Managing Cloud Tiers
The way cloud providers set up their tiers of storage presents IT with a unique challenge and a unique opportunity. IT can get the reward of significant cost savings if they can figure out what data should go on a specific storage tier. But if they choose wrong they could cost the organization a lot of money in access fees.
While almost every cloud application could leverage multiple tiers, the two most obvious are cloud backup solutions and cloud archiving solutions. Each of these solutions manage data at different stages of its life cycle. When data first gets to the cloud, the chances of it being restored, recalled or used in some other way is fairly high.
The performance of the storage that data is stored on may matter given the circumstance. For example, backup and recovery solutions that want to instantiate an application in the cloud will want to have fairly high performing storage. Even in an archive use case, there may be a need for high-performance storage. For example, if the archive has audio and/or video data in it the organization may want to leverage a cloud service to transcribe that data.
In both the backup and archive use cases though, as data ages, the chances of it being accessed again continues to decrease. At that point, it makes sense to store it on colder storage that is cheaper per-GB even though there is a high charge to access it.
A key cloud tier management is to have a software solution that can take advantage of the various tiers. If IT has a solution that only supports one tier of storage, getting data to the other tiers is a manual, time consuming process. And in some cases, moving data outside of the purview of the application might cause problems for the application. IT planners need to be careful. Some software solutions will support more than one tier, but they can’t move data between tiers.
The second requirement is for the solution to analyze the data that it is responsible for and move that data to the appropriate cloud storage tier as the data ages. The solution should also provide additional insight and analytics so IT planners can make their own decisions about where to place cloud data.
Finally, the solution should be able to index data across tiers but store the actual index in one of the more active tiers. A separate metadata index enables search to occur against data in the cold storage tier without it actually being accessed.
The search function itself should be sensitive to the cloud storage paradigm. First, it should allow various levels of granularity, which allows critical sensitive data to have a full context level search table build against it while other data is only tracked by more simple metadata like access date, file name and owner.
Cloud tiers, when used correctly, can save the organization hundreds of thousands of dollars in cloud storage costs. But IT has to use those tiers carefully because of access charges. Cloud storage management software should manage the data it sends to the cloud, automatically move it between tiers as needed and provide IT insight so it can make its own decisions on data placement.
To learn more about managing cloud tiers and how to fully leverage the cloud for archiving, watch our on demand webinar “Cloud Archiving – Amazon Glacier, Microsoft Azure or Something Else?”