Overall Satisfaction with Amazon Glacier
We used Amazon Glacier to store backups of virtual machine images. A virtual machine image may take gigabytes and, considering that we made snapshots quite often, that would generate a lot of data very fast. Amazon Glacier turned out to be a cheap solution to store all that data.
- Cheap storage of backup data.
- Can be used as a part of the entire suite of tools from Amazon, without requiring you to leave the familiar stack.
- Enumeration of data takes hours. In order to get data back you'll need to do inventory of the bucket, but that takes a few hours to complete.
- Data and bucket deletion also requires inventory. Moreover, it is not possible to delete a bucket that has been recently written to - making bucket deletion a lengthy ordeal: first, you perform inventory, then delete your files, but you cannot delete the bucket as you recently deleted files, so you have to wait for another day in order to delete the bucket itself.
- While data storage is cheap and easy, data retrieval can be expensive as you only get a download quota that is a few percent of your storage.
- Cheap backup storage - up to the level where it becomes only tiny fractions of your entire project, whereas other solutions would be much more expensive.
- The established reliability of AWS also makes you sleep better at night knowing your data is going to be OK.
It is significantly cheaper than other services, however, it is because it actually is a slightly different service. The other services we've tried allow live reading/writing of data as needed, whereas Glacier is a "cold storage" service. So essentially your choice ends up being based on what you plan to do with your data: keep it indefinitely or retrieve and modify it (potentially often and multiple times).
Using Amazon Glacier
Pros | Cons |
---|---|
Easy to use Technical support not required Well integrated Consistent Quick to learn Feel confident using Familiar | Do not like to use |
- Data storage
- Data retrieval, deletion and modification