Welcome to the October 2023 edition of the pganalyze newsletter!
We've been busy working on the next generation of pganalyze Index Advisor, enabling better recommendations and more control over what gets prioritized, using our new constraint programming model for index selection, first presented at PGCon earlier this year.
Join our webinar on November 15th to learn all about how the new model works, how you can participate in early access for the updated version of pganalyze Index Advisor, upcoming recommendations for index consolidation (sometimes one index is better than two!), new controls for reducing indexing impact to Postgres' HOT updates, and more.
You can also find our recent podcast interviews and "5mins of Postgres" episodes later in this newsletter. And last but not least, I'm excited to share our newly opened Head of Sales position!
In this webinar you will get an introduction to our new Index Selection model, based on Constraint Programming and CP-SAT, that can automatically determine which set of indexes to create for a given Postgres query workload, based on objectives chosen by the developer or DBA.
This new model forms the basis of the next generation of pganalyze Index Advisor, and in this webinar we'll give you a sneak peek of upcoming functionality, such as the ability to consolidate existing indexes, controls for considering the impact of indexing to Postgres' HOT Updates, and the ability to prioritize indexing choices for user-facing queries.
We hope to see you there. Feel free to share the link and to invite your peers!
pganalyze is growing and becoming the standard way to tune and monitor Postgres databases at any scale, with over 500 customers, small and large. We are looking to grow our team with a knowledgeable and experienced Head of Sales. Learn more on our careers page.
We look at improvements to async queries with find_by_sql, composite primary keys, compiling of common table expressions, faster test suites, and more.
We walk through pgbouncer's transaction mode, performance impact of prepared statements, and caveats when using pgbouncer 1.21 with prepared statements in transaction mode.
Subscribe on YouTube
We publish weekly videos on Postgres, our conference talks, tutorials, product walk-throughs, and more.
Postgres 14+: Include toplevel attribute in statement statistics key
This could have caused statistics to be incorrect in Postgres 14+ when the same query was called both from within a function (toplevel=false) and directly (toplevel=true), with pg_stat_statements.track set to "all"
If affected, the issue may have shown by bogus statistics being recorded, for example very high call counts, since the statement stats diff would not have used the correct reference
OpenTelemetry integration
Support sqlcommenter format query tag (key='value') for traceparent
Add a new config setting otel_service_name / OTEL_SERVICE_NAME for customizing the OpenTelemetry service name
Other improvements and bugfixes
Query Performance: Improve guidance when a query has no query samples
Exclude Google Cloud SQL setting from "Config - Disabled features" check
VACUUM Advisor xmin horizon check: Correctly handle servers with stale statistics data
Connection Tracing: Adjust logic that fetches log lines for a PID to match logic for "Log Entries" count
Review us on G2
Please leave your feedback (good or bad!) in this survey - we appreciate it!
