> ## Documentation Index
> Fetch the complete documentation index at: https://docs.paradedb.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Performance Tuning

> Several settings can be tuned to improve the performance of aggregates in ParadeDB

### Configure Parallel Workers

ParadeDB uses Postgres parallel workers. By default, Postgres allows two workers per parallel query.
Increasing the number of [parallel workers](/documentation/performance-tuning/reads) allows parallel queries to use all of the available hardware on the host machine and can deliver significant
speedups.

### Run `VACUUM`

`VACUUM` updates the table's [visibility map](https://www.postgresql.org/docs/current/storage-vm.html),
which speeds up Postgres' visibility checks.

```sql theme={null}
VACUUM mock_items;
```

If the table experiences frequent updates, we recommend configuring [autovacuum](https://www.postgresql.org/docs/current/routine-vacuuming.html).

### Run `pg_prewarm`

The `pg_prewarm` extension can be used to preload data from the index into the Postgres buffer cache, which
improves the response times of "cold" queries (i.e. the first search query after Postgres has restarted).

```sql theme={null}
CREATE EXTENSION pg_prewarm;
SELECT pg_prewarm('search_idx');
```