The following are the steps and guide in how to install and setup Vectorchord extension in your EDB EPAS PostgreSQL cluster.

First, you will need to install the two RPM packages for PgVector and VectorChord from EDB Latest package repositories.

 

Execute the following command to check that two packages are installed on your redhat operating system:

 

rpm -qa ‘*edb*’ | grep vect

edb-as16-pgvector0-0.8.2-1.el8.x86_64

edb-as16-vectorchord-1.1.1-1.el8.x86_64

After that, you will need to install both extensions as “admin” and set it up as follows:

postgres=# ALTER SYSTEM SET shared_preload_libraries = “vchord”;

 

postgres=# CREATE EXTENSION IF NOT EXISTS vector;

 

postgres=# create extension vchord;

 

Then, reboot the whole database instance cluster and then verify that extensions were installed successfully by executing the following:

postgres=# \dx

                                                      List of installed extensions

        Name        | Version |   Schema   |                                         Description

——————–+———+————+———————————————————————————————

edbspl             | 1.0     | pg_catalog | EDB-SPL procedural language

pg_stat_statements | 1.10    | public     | track planning and execution statistics of all SQL statements executed

plpgsql            | 1.0     | pg_catalog | PL/pgSQL procedural language

vchord             | 1.1.1   | public     | vchord: Vector database plugin for Postgres, written in Rust, specifically designed for LLM

vector             | 0.8.2   | public     | vector data type and ivfflat and hnsw access methods

(5 rows)

After that perform testing at a user-database level (ensure the 2 extensions are created at database level):

postgres=# \c dummy

You are now connected to database “dummy” as user “enterprisedb”.

dummy=# \dx

                                                 List of installed extensions

  Name   | Version |   Schema   |                                         Description

———+———+————+———————————————————————————————

edbspl  | 1.0     | pg_catalog | EDB-SPL procedural language

plpgsql | 1.0     | pg_catalog | PL/pgSQL procedural language

vchord  | 1.1.1   | public     | vchord: Vector database plugin for Postgres, written in Rust, specifically designed for LLM

vector  | 0.8.2   | public     | vector data type and ivfflat and hnsw access methods

Now, create a table and insert a dummy data for verification and simulation:

CREATE TABLE items (id bigserial PRIMARY KEY, embedding rabitq8(3));

CREATE INDEX ON items USING vchordrq (embedding rabitq8_l2_ops);

INSERT INTO items (embedding) VALUES (quantize_to_rabitq8(‘[0,0,0]’::vector));

INSERT INTO items (embedding) VALUES (quantize_to_rabitq8(‘[1,1,1]’::vector));

INSERT INTO items (embedding) VALUES (quantize_to_rabitq8(‘[2,2,2]’::vector));

SELECT id FROM items ORDER BY embedding <-> quantize_to_rabitq8(‘[0.9,0.9,1.1]’::vector) LIMIT 100;