PostgreSQL 18 Vector Database with pgvectorscale on Ubuntu 24.04 LTS

Azure Artificial Intelligence (AI)

A vector search database: PostgreSQL with pgvector and pgvectorscale, adding StreamingDiskANN approximate nearest neighbour search over your embeddings.

Base

Hardened build

minimal ports, security patches applied at build time

Access

Unique credentials

generated on first boot, readable only by root

Verified

Boots working

services pass a health gate before release

Support

24/7, 365 days

by email and live chat, 24 hour response SLA

Overview

This appliance is a vector search database built on PostgreSQL, with both open source vector extensions enabled side by side. pgvector supplies the vector column type, the distance operators and the HNSW and IVFFlat index families. pgvectorscale adds StreamingDiskANN, a disk resident approximate nearest neighbour index inspired by Microsoft research, together with statistical binary quantisation, so a large embedding collection can be served from a modest machine rather than held entirely in memory.

Because it is still PostgreSQL, embeddings live next to ordinary relational data and you keep transactions, joins, SQL and the whole driver and tooling ecosystem. It suits retrieval augmented generation over your own documents, semantic and hybrid search at scale, recommendation and deduplication, and any workload where similarity ranking has outgrown a plain vector column but a separate specialist database would be a second system to run.

Why the cloudimg image

pgvectorscale has no distribution package: it is a Rust extension that has to be compiled. cloudimg compiles it from the pinned upstream source, verifies the licence at that exact tag, then removes the compiler so the image ships the extension and nothing else. Both extensions arrive already enabled in a ready to use database with a demo collection and a StreamingDiskANN index, so a real similarity query runs the moment the machine boots. There is no default password to change and no shared database: every instance initialises its own database cluster on first boot with its own generated passwords and its own TLS certificate, written to a file only the root user can read, and the database listens on loopback until you choose to open it. The paired deployment guide covers indexing your own embeddings and confirming the planner uses the index, and every instance carries 24/7 support.

Common uses

Retrieval augmented generation and semantic search over large document collections
Approximate nearest neighbour search on embeddings that no longer fit comfortably in memory
Recommendation, deduplication and similarity ranking alongside relational data

Overview

Why the cloudimg image

Common uses

More in Artificial Intelligence (AI)

Argilla on Ubuntu 24.04 LTS

Bifrost on Ubuntu 24.04

ContextForge MCP Gateway on Ubuntu 24.04 LTS

Crawl4AI on Ubuntu 24.04

DeepWiki-Open on Ubuntu 24.04 LTS

Dify on Ubuntu 24.04 LTS