Archive Β· Β· 3 min read

Amazon S3 Glacier Re:Freezer

AWS Labs has a lot of open source code up on GitHub. This repo helps you visualize your AWS workloads and highlight areas of cost.

Amazon S3 Glacier Re:Freezer

AWS Labs has a lot of open source code up on GitHub. Amazon S3 Glacier Re:Freezer is one of those projects.

It’s got a simple premise: get the data out of Amazon S3 Glacier and back into Amazon S3.

That sounds simple but the solution is surprisingly nuanced. I call out a few more details in the Twitter thread below…

Tweet 1/10 πŸ‘‡ Next tweet

let's take a look at the Amazon S3 Glacier Re:Freezer project from @awscloud Labs the project is available at 🧡☁️ #cloud #devops

Tweet 2/10 πŸ‘‡ Next tweet πŸ‘† Start

this thread can be see unrolled at the previous thread about AWS Perspective is up at 🧡☁️ #cloud #devops

Tweet 3/10 πŸ‘‡ Next tweet πŸ‘† Start

the goal of this project is simple: get data out of @awscloud S3 Glacier and back into an Amazon S3 bucket that shouldn't be too hard... 🀣🀣🀣 🧡☁️ #cloud #devops

Tweet 4/10 πŸ‘‡ Next tweet πŸ‘† Start

I laugh because this is the solution architecture πŸ‘‡ 🧡☁️ #cloud #devops

Tweet 5/10 πŸ‘‡ Next tweet πŸ‘† Start

think about that for a minute. it builds out resources in SEVEN different @awscloud services just to move data from one service to another worse, those two services are intertwined already! if there was ever a solution crying out to be in product, it's this 🧡☁️ #cloud #devops

Tweet 6/10 πŸ‘‡ Next tweet πŸ‘† Start

ok, rant aside (for now), this is a very useful solution. Glacier is the long term, cold storage ( where writing is easy and reading is slow and expensive Glacier is the "just in case" place for storage. 🧡☁️ #cloud #devop

Tweet 7/10 πŸ‘‡ Next tweet πŸ‘† Start

Glacier is a great service because it's super low storage costs ( make it easy to keep things "available" you're charge to get data out and for those times you actually need that data, it's usually worth it 🧡☁️ #cloud #devop

Tweet 8/10 πŸ‘‡ Next tweet πŸ‘† Start

you would think that restoration process is automation in a slick manner. it isn't natively this solution does a nice job of smoothing out this process simply: - point at the Glacier Vault - pick your retrieval tier - target bucket - target storage class 🧡☁️ #cloud #devop

Tweet 9/10 πŸ‘‡ Next tweet πŸ‘† Start

the complexity in the architecture πŸ‘† is because the data retrieval from @awscloud Glacier is glacially slow like a herd of 🐒🐒🐒 who aren't sure which way they are supposed to be going type of slow 🧡☁️ #cloud #devop

Tweet 10/10 πŸ‘‡ Next tweet πŸ‘† Start

all of the different components in the design make sure that the data gets to the target bucket with a high level of reliability should this be part of the core service offering ? yes but it's not, so use this code /🧡☁️ #cloud #devop

Read next