AWS has recently announced that Amazon Aurora supports major version 13 of PostgreSQL. The compatibility with the latest major PostgreSQL version adds improvements such as higher performing partitioned table queries and deduplication of B-tree index entries.
Other benefits and enhancements of running PostgreSQL 13 are incremental sorting to accelerate data sorts, parallel processing of indexes with the VACUUM command and new security capabilities. The release adds support for bool_plperl, an extension to write procedures in Perl, and includes new ways to monitor activity within a database, using query sampling, improved WAL usage statistics and new system views.
The added support for the latest PostgreSQL version comes a few months after the announcement of supporting PostgreSQL 12 on Amazon Aurora and of PostgreSQL 13 on Amazon RDS. Jean Baro, senior principal solution architect at Questrade Financial Group, comments:
It's interesting to watch AWS shorten the release cycle of Aurora PostgreSQL versions. For version PG 12, they took 18 months to catch up with the community version. For version PG 13, it took 11 months. I hope they use more and more of the Open Source PostgreSQL version in the future. It would be beneficial to everybody. I also know AWS is contributing to PostgreSQL, just like Microsoft is doing. Great times for PostgreSQL.
Shayon Sanyal, senior database specialist solution architect at AWS, recently presented a webinar about PostgreSQL 13 and the options to run it on AWS. In a separate article called "Designing high-performance time series data tables on Amazon RDS for PostgreSQL", Jim Mlodgenski, principal database engineer at AWS, and Andy Katz, senior product manager at AWS, explain how table partitioning and partition pruning are more effective in PostgreSQL 13, a critical feature for time series data tables:
By their nature, time series tables grow continuously. When these tables get large, splitting these tables into many smaller tables will improve database performance and make maintenance much easier. (...) One of the key features in PostgreSQL 13 enabling this scale is an improvement in partition pruning. (...) Earlier versions of PostgreSQL were not as effective in pruning partitions, requiring many queries to scan all partitions.
AWS is not the only cloud provider offering a managed PostgreSQL 13. Version 12 and 13 have been recently added to Azure Database and Google Cloud SQL for PostgreSQL supports version 13 as well as older ones.
To match updates by the open source community to older PostgreSQL major versions, Amazon Aurora has also announced compatibility and support for 9.6.22, 10.17, 11.12, and 12.7. There is no difference in pricing according to the version used.