Hi I wanted to understand how well HyperDex scales with very large datasets much bigger than RAM. I'm also interested in knowing how HyperDex would compare with FoundationDB and CockroachDB in terms of raw performance as well as scalability when you consider a Multi Data Center scenario. Have there been any tests done for the same ?