EPIC: Archive workflow support

[dneufeldcu]

Summary

As a	a data archive
I want to be able to	preserve data for the long-term at a reasonable cost
So that I can	feel confident in cloud based archival processes

Description

Move files landing in s3 to glacier based on temporal and parentId rules. Be able to restore the data to s3. Understand any AWS limits.

Tasks

• [x] Design session and documentation of discussion
• [x] Know when a file goes into glacier
• [x] Persist the identifier and location of the file
• [x] Be able to retrieve the file from glacier

Intelligent Storage

• https://docs.google.com/document/d/1ga10PGamia4Y7-Hl4qWFNJbzr-QlGuPilBNv14y6SDg/edit#heading=h.bh4kpw8i4ph6

[caesterlein]

relevant - https://docs.google.com/document/d/1nxYJD4tk6yfT3GdGNHCR8lH8lfYRO4C364ZrJ68NVVk/edit?usp=sharing

[dneufeldcu]

Currently we are using the 'glacier' resource type in boto3. This should be switched to s3 resource as glacier api will be deprecated.