Is DocumentDB Really PostgreSQL?

On Thursday I wrote about the new MongoDB compatible DocumentDB from AWS and its potential impact on MongoDB going forward. Following up from that, a colleague pointed me to a very interesting thread on Y Combinator. That thread makes the case that, based on a number of clues, it appears that DocumentDB may actually be based on PostgreSQL, and in particular Aurora PostgreSQL compatible edition!  

Some of the technical clues include:  

• DocumentDB scales vertically for writes and horizontally for reads via replication, just like PostgreSQL. From the AWS DocumentDB FAQs, “Amazon DocumentDB's compute capacity can be scaled up by creating larger instances and horizontally (for greater read throughput) by adding additional replica instances (up to 15) to the cluster.”
• Also from the FAQs, “Amazon DocumentDB automatically maintains six copies of your data across three Availability Zones”, again, just like Aurora PostgreSQL.
• Identifiers are limited to 63 characters; the same as PostgreSQL as noted by the AWS documentation on DocumentDB Limits.
• This document also notes that a collection is limited to 32TB in size, coincidentally (or not) the maximum size of a table in PostgreSQL.  
• DocumentDB also does not allow the UTF-8 null character as noted functional differences section of the DocumentDB Developer Guide.

Some of this evidence may be circumstantial. Obviously, PostgreSQL isn’t the only DBMS that scales writes vertically and reads horizontally via replication, but when you add this all up, especially some of the specific limitations, I think it makes a pretty compelling argument that PostgreSQL is the engine powering AWS DocumentDB. And since PostgreSQL is also at the heart of RedShift and is one of the most popular engines for RDS and Aurora, it is clear that PostgreSQL is one of the major engines driving Amazon.  

BTW, one of my colleagues here at EnterpriseDB and leader in the PostgreSQL community, Robert Haas also shared some of his thoughts on the news about DocumentDB on his blog.  Definitely worth checking out.