Interested in how the pganalyze Indexing Engine works?
Read the technical deep-dive blog on the pganalyze Indexing Engine, or continue reading for the main benefits.
Read the technical deep-dive blog on the pganalyze Indexing Engine, or continue reading for the main benefits.
A better approach for indexing your Postgres database
pganalyze Indexing Engine automatically finds the right indexes for your database, through a three step approach:
- Break down your query workload into individual scans of the tables involved
- Assess each table to determine the optimal index configuration
- Surface recommendations together with schema and query statistics in the pganalyze Index Advisor
"What If?" analysis with zero production impact
- The pganalyze Indexing Engine runs fully within the pganalyze app, based on query and schema statistics
- Detailed analysis compares all possible indexes for a specific query and its scans
- No special extensions are required to enable the Indexing Engine for your production database
The "Good Enough" index selection algorithm
- The pganalyze Indexing Engine recommends the optimal set of indexes that covers all queries
- Recommendations are balanced between the improved query performance, and the added Index Write Overhead
- This result is deterministic - when the same version of the Indexing Engine runs again with the same inputs, it produces the same recommendations
Fully integrated into pganalyze
The new version of the pganalyze Index Advisor is powered by the pganalyze Indexing Engine, and provides detailed recommendations within the context of your schema and query statistics.
Simply set up pganalyze for your production system, and it will automatically collect necessary data that is used to find missing indexes for your query workload.