A Look at Amazon Glacier

A Look at Amazon Glacier

September 17

Amazon Web Services has recently released a new data backup, storage and archival service called Glacier. This service promises $0.01 per GB storage, but there are some things you need to know before you consider it.

I’ll begin by explaining that AWS uses Glacier to store data “that is not likely to be accessed.” This means it’s an archive, and as such works well enough for records you may not need for some time. The name “Glacier” puts things in perspective as data is, in a way, frozen. Let’s take a closer look at how Glacier works:

You can put data in Glacier for a penny per gigabyte per month. That is awesome. If there is something you decided you don’t want in there, you can delete it but, if you delete it from Glacier before the first ninety days, you’ll still have to pay a prorated price until the ninety days is up.

Now, suppose three months have gone by and you decide you need tax records proving the Star Trek inflatable captain’s chair is a legitimate business purchase. You can submit a job request and in 3-5 hours you will be able to begin downloading. If the data you need exceeds 5% of the total data stored, you must pay a retrieval fee for anything you want over your 5% monthly allowance.

Pricing gets more complicated when your monthly bill comes. The volume of storage billed per month is based on average storage of that month, measured in something called gigabyte-months. After Amazon adds extra data space for indexing and metadata, and the “TimedStorage-ByteHrs” for the month (determined by how many bytes per hour, per day, multiplied by the amount of days that month), you will have your total (I think).

If you’re like me, at this point your brain will likely be as frozen as the data in the Glacier—it’s difficult to accurately determine what your monthly bill will be with this pricing structure, but if you take the time to figure it out, Glacier does have advantages.

While these examples may make Glacier seem threatening, if you understand the pricing structure and need a simple, cheap place to store things you won’t need for a while, freezing it in the Glacier is comparatively inexpensive and secure.

While you should probably not put data that might be time-critical in the Glacier, you may find you need something now and not five hours from now. In the event of an emergency, data retrieval time is absolutely critical—you can’t simply submit a job request and wait for a frozen old data librarian to find your data and deliver it to you, so take care deciding which data to freeze in the Glacier.

Does Glacier Work with ShadowProtect?

Currently, StorageCraft ShadowProtect does not have functionality with Glacier. Amazon explains on Glacier’s info page,

“Amazon Glacier provides a management console. You can use the console to create and delete vaults. However, all other interactions with Amazon Glacier require programming. For example, to upload data, such as photos, videos, and other documents, you must write code and make requests using either the REST API directly or the AWS SDK for Java and .NET wrapper libraries.”

Glacier requires programming for many additional functions, and doesn’t have an end user interface, so it would take quite a bit of work to make our products function with Glacier well enough to meet our standards. Since Glacier is new, we’re still learning about it and haven’t determined how it fits into our roadmap.

As far as cloud backup and recovery needs are concerned, you can look forward to the upcoming StorageCraft Cloud Services this fall. If you have an emergency, you can access your data and even spin up a virtual machine in the cloud and business can resume in a matter of minutes.

If you need a place to store data that you won’t need for some time and recovery time is not an issue, you can freeze it in the Glacier, but if you need a reliable and fast backup and recovery solution, StorageCraft Cloud Services will help you keep the flame of your business going.