pgvector v0.1.0

Databases

PostgreSQL + pgvector — a ready-to-use PostgreSQL server with the pgvector extension for storing embeddings and running vector similarity search right in your relational database. Ideal as the retrieval store for RAG and semantic search. Deployed as a host-networked Nomad service with a persistent data volume.

nomad-pack run pgvector --registry nomploy

…or one line (add registry + run)

curl -fsSL https://packs.nomploy.com/install.sh | sh -s -- pgvector

Needs nomad-pack on PATH. The script only adds the nomploy registry and runs this pack.

Source ↗ Project ↗ ★ 23.1k ⚑ Report an issue

1 task db 5432 1 volume image pgvector/pgvector:pg16 pinned :pg16 image bumped today

Variables 11

values.hcl

Save as values.hcl, edit, then run:

nomad-pack run pgvector -f values.hcl --registry nomploy

# The name of the Nomad job.
job_name = "pgvector"

# The Nomad namespace to deploy into.
namespace = "default"

# The datacenters to deploy to.
datacenters = ["*"]

# The pgvector PostgreSQL image (postgres + pgvector). Match the tag to your Postgres major version. Pin a tag in production.
image = "pgvector/pgvector:pg16"

# Host port for PostgreSQL.
port = 5432

# Initial database created on first boot (POSTGRES_DB).
db_name = "app"

# PostgreSQL superuser (POSTGRES_USER).
db_user = "postgres"

# PostgreSQL password (POSTGRES_PASSWORD). CHANGE THIS.
db_password = "change-me-please"

# Named volume for PostgreSQL data (/var/lib/postgresql/data). Holds all databases and vectors.
data_volume = "pgvector_data"

# Placement constraints. Pin to the node holding the volume. On a nomploy cluster: attribute = "$${meta.nomploy_control_plane}", operator = "=", value = "true".
constraints = []

# Resources for the PostgreSQL task.
resources = {
    cpu    = 500
    memory = 512
  }

Name	Type	Default	Description
job_name	string	"pgvector"	The name of the Nomad job.
namespace	string	"default"	The Nomad namespace to deploy into.
datacenters	list	["*"]	The datacenters to deploy to.
image	string	"pgvector/pgvector:pg16"	The pgvector PostgreSQL image (postgres + pgvector). Match the tag to your Postgres major version. Pin a tag in production.
port	number	5432	Host port for PostgreSQL.
db_name	string	"app"	Initial database created on first boot (POSTGRES_DB).
db_user	string	"postgres"	PostgreSQL superuser (POSTGRES_USER).
db_password set me	string	"change-me-please"	PostgreSQL password (POSTGRES_PASSWORD). CHANGE THIS.
data_volume	string	"pgvector_data"	Named volume for PostgreSQL data (/var/lib/postgresql/data). Holds all databases and vectors.
constraints	list	[]	Placement constraints. Pin to the node holding the volume. On a nomploy cluster: attribute = "$${meta.nomploy_control_plane}", operator = "=", value = "true".
resources	object	{ cpu = 500 memory = 512 }	Resources for the PostgreSQL task.

Back up this pack

This pack stores data in one Docker named volume: pgvector_data

⚠ This pack bundles a database. A cold copy of the volume can be inconsistent — for a reliable backup, dump the DB (pg_dump / mysqldump) or stop the job while backing up.

restic

# Run on the node hosting this pack. Point restic at your repo first:
#   export RESTIC_REPOSITORY="s3:https://<account>.r2.cloudflarestorage.com/<bucket>"
#   export RESTIC_PASSWORD="<repo-password>"
#   export AWS_ACCESS_KEY_ID=<key>  AWS_SECRET_ACCESS_KEY=<secret>
restic backup \
  /var/lib/docker/volumes/pgvector_data/_data

rclone (sync to S3/R2)

rclone sync /var/lib/docker/volumes/pgvector_data/_data backup:<bucket>/pgvector_data

Paths assume the default Docker volume location (/var/lib/docker/volumes). Restore by stopping the job, restoring files into the same volume, and re-running the pack.

Readme

pgvector

pgvector on PostgreSQL — a ready-to-use Postgres server with the pgvector extension for storing embeddings and running vector similarity search directly in your relational database. Keep your app data and your vectors in one place — ideal as the retrieval store for RAG and semantic search.

Single host-networked Nomad service with a persistent data volume.

Deploy

nomad-pack registry add nomploy https://github.com/Nomploy/nomad-packs
nomad-pack run pgvector --registry=nomploy

Configure

Variable	Default	Description
`port`	`5432`	PostgreSQL port.
`db_name`	`app`	Initial database (`POSTGRES_DB`).
`db_user`	`postgres`	Superuser (`POSTGRES_USER`).
`db_password`	`change-me-please`	Password (`POSTGRES_PASSWORD`). Change it.
`data_volume`	`pgvector_data`	`/var/lib/postgresql/data` — all databases and vectors.
`image`	`pgvector/pgvector:pg16`	Image (match the tag to your Postgres major version). Pin in production.
`resources`	`{ cpu = 500, memory = 512 }`	Task resources.

Enable the extension once per database, then use vector columns:

CREATE EXTENSION IF NOT EXISTS vector;
CREATE TABLE items (id bigserial PRIMARY KEY, embedding vector(1536));
-- nearest neighbors: ORDER BY embedding <-> '[...]' LIMIT 5;

A drop-in vector store for the ollama / open-webui / anythingllm packs. Serves an unencrypted SQL port — keep it on an internal network. Pin the job to the node holding the volume with constraints.