Hi, and welcome to cloudacademy.com's video series on data management using the Amazon Cloud. This video we're going to discuss Managing Glacier archives through Amazon's high level API and specifically using Java. Glacier archives are very cheap about one cent per gigabyte per month, the trade-off as we mentioned before is that there's a very high latency can be hours after an upload for instance before you can access the data and it could be hours before a download is initiated. Also the archives on Glacier can only be accessed and managed via a program API, however given those restrictions Glacier can be a terrific platform for storing less important or data that's important but doesn't need immediate access. Let's take a look at a sample code that will upload an archive from your instance or from your computer anywhere to a Glacier vault. You will already have created a big Glacier vault, we'll assumed that you had no trouble doing that and you're like giving it a unique name and it's important to remember that the name you give your vault is in fact unique. So the first in this case highlighted selection of the code is the imports that is those libraries that are installed with the AWS toolkit for Eclipse, if you happen to be using Eclipse. These libraries are what actually runs your code under the surface, the next string you will have to populate is the vault name just so that your code knows where to look.

The vault name in this case is "myuniquename876" this is just a vault I created in the previous video for demonstration purposes. You'll next populate the string archive to upload, so the program will know which archive actually is meant to be moved up to Glacier. It has to be the archive name in its absolute location, that is in this case "home/awscontrol" which would be the name of a user account in your system whether it's in a Amazon instance or on your local computer. Just I made up the name "awscontrol" it doesn't have to be that name obviously, and it might be an archive which is compressed using tar and Gzip called "mydata" Your code will have to provide credentials, this is done through the profile credentials provider variable which is populated by the file credentials. This file is usually created and stored in the home directory of the user who created it in a hidden directory called aws, .aws means it's a hidden directory in a file called credentials.

There you will replace your access key ID and your secret access key with the values that you can get from your Amazon dashboard. Finally, the code actually does the task it was meant to achieve and that is first by establishing a connection to the Amazon vault you've created through using your credentials and then actually uploading the archive you created and designated to the vault you've designated, that's how we upload.

Downloading archives in Glacier vault is a little different, you'll notice that there are different libraries imported before this code can be run, you'll notice also that you'll need the archive ID that is the ID that Amazon gave your archive when you uploaded it through your Glacier vault. We'll need to insert that to populate the string archive ID you'll also need to provide the absolute download file path. In our case it's "home/awscontrol/Downloads" one additional option, which is applicable both to uploads and downloads is the end point which Amazon region is hosting your vault.

By default the region is set to "us-east-1" if however your vault is hosted in a different region then you can populate the Glacier client set end point value. In this case glacier.us-west-2 if that happens to be the location of your vault.