Add OAM Bucket for AWS Open Data Hosting

[cgiovando]

AWS has offered to include OAM imagery in their Open Data Program. This provides an opportunity for OAM to have imagery hosted in a sponsored storage space, reducing the overall costs of running the project, and increasing visibility and accessibility to the data.

In the Appendix A of the Onboarding Handbook AWS provides guidance on how the bucket with data should be organised. Unfortunately the recommended structure does not reflect how the data is currently stored in the oin-hotosm bucket where all OAM imagery is hosted. It would take a significant re-writing of the OAM code to map the current structure to the one proposed by AWS. This could be considered in future versions of OAM. For now we understand that it's still possible to use the current OAM bucket structure, even if it doesn't fit the recommended schema.

During the last OAM stakeholders meeting we discussed the possibility of adding the AWS Open Data hosted bucket to the list of indexed storage endpoints that are indexed by the OAM Catalog. The same bucket would then be used by the OAM Uploader to upload new images. The existing imagery could be moved to the new AWS Open Data bucket, but it probably will need to also stay in the current bucket to avoid breaking any existing links on the Internet.

These should be the logical steps to implement the above - please comment if I'm missing anything or if you would suggest using a different approach:

• Review section 5 of the Onboarding Handbook with instructions on the required bucket configuration
• Decide the new bucket name and create it within the AWS hosted account
• Add the new bucket to the list of endpoints indexed by the OAM Catalog
• Change the OAM Uploader configuration to use the new AWS Open Data bucket
• Copy all existing OAM data from the current bucket to the new one
• Test on staging
• Deploy to production

Kindly asking @dakotabenjamin and @Komzpa to review and provide guidance for implementation.