I came across an interesting library on The Changelog
– a proxy for S3 metadata called mimicdb
by Nathan Cahill. This piqued my interest because one of our CMS implementations is able to serve its assets from S3, but we have noticed that the time to look up asset metadata can be expensive, especially when the server making the request isn’t within Amazon’s ecosystem.
The library mirrors metadata locally so the developer can opt in to using cached data, rather than making a full trip to S3 and back – representing a significant saving on time and cost for applications that work with thousands of assets.