Elasticsearch quality Full Text Search on Postgres: Managed ParadeDB on Ubicloud

October 1, 2024 · 5 min read
Umur Cubukcu
Co-founder / Co-CEO

At Ubicloud, we're building an open and portable cloud. Our goal is to provide the highest performing cloud services at the lowest cost. Today, we're excited to announce our partnership with ParadeDB to bring you a fully managed ParadeDB – Postgres for Search and Analytics service, by combining the Ubicloud-managed PostgreSQL service with ParadeDB's powerful pg_search and pg_analytics Postgres extensions.

Read on to learn more about how Elasticsearch-quality full-text search is now possible natively with PostgreSQL.

The Problem

We've seen too many teams hit with the complexity and cost of running their database and search infrastructure in two separate systems. One of the most typical setups involves PostgreSQL for the database and Elasticsearch for search. This dual-approach leads to multiple challenges:

  • Increased operational overhead from managing two systems. One common pattern for this is running UPDATE-heavy workloads in Postgres, while denormalizing the data for Elastic search.
  • Complex data synchronization between the database and search engine; including risk of data inconsistencies
  • Higher costs from running and maintaining multiple services

As cloud services have evolved, the solutions offered by major providers have become more complex and expensive. Self-hosting all the different components is tricky.

The Solution

With ParadeDB on Ubicloud, we take a different approach:

  • ParadeDB transforms PostgreSQL into a performant search and analytics engine, and is packaged as a PostgreSQL extension.
  • Ubicloud PostgreSQL runs the ParadeDB extension in the cloud with automated backups, point-in-time-recovery, high-availability, and all the ease of use that comes with managed PostgreSQL.
  • Your primary PostgreSQL database remains unchanged, whether it was already running on Ubicloud, or managed on a different cloud. A new replica of it running ParadeDB is created, and gets managed by Ubicloud.
  • High-speed NVMe disks on Ubicloud maximize your search performance -- sometimes by 10x, or more.
  • When you have a question about ParadeDB, you get immediate and direct access to the ParadeDB team.

And of course, all of this is implemented with open source.

Key features and performance

Here are some of the features you can expect:

  • PostgreSQL functionality and ACID compliance, with immediate searchability of data after writes.
  • Powerful search capabilities, including BM25 scoring, custom tokenizers, and hybrid search.
  • Fast analytics and faceting support.
  • Full JOINs support with full-text search queries, including recursive JOINs.
  • Automatic backups and point-in-time recovery.
  • High availability with automated failovers.
  • Encryption at rest and in transit.
  • Dedicated cores powering each instance, NVMe disks for maximum analytics performance.

Performance is a key focus, driven both by how ParadeDB extends PostgreSQL and how the Ubicloud infrastructure enables the workload. Initial benchmarks from the ParadeDB team show results often 5x faster on instances with comparable costs; and this blog post outlines the performance characteristics of ParadeDB at Alibaba.

Best value for your database spend

Our fully managed ParadeDB service starts at just $0.193 / hour, making it much more cost-effective than moving data from PostgreSQL to a separate Elasticsearch instance using traditional cloud providers. This pricing includes:

  • 2 dedicated vCPUs, 8 GB RAM, and 128 GB disk.
  • Full ParadeDB functionality.
  • Ubicloud's managed service benefits.

Pricing scales linearly as you grow, up to 60 vCPUs | 240 GB RAM | 4TB disk, with billing in minutely increments for maximum flexibility.

Top-notch support directly from the creators

Unlike all earlier approaches before it, ParadeDB on Ubicloud goes much beyond simply enabling the ParadeDB extension on Ubicloud PostgreSQL. When you provision a ParadeDB PostgreSQL instance on Ubicloud, you get direct access to the team that built the ParadeDB product. This unique approach ensures:

  • Direct communication channels with ParadeDB developers while your PostgreSQL remains managed by the Ubicloud team.
  • Personalized advice on optimizing your ParadeDB implementation.
  • Access to the latest ParadeDB features and updates.

And for everything related to running PostgreSQL smoothly in the cloud, you get the Ubicloud team with all our years building core PostgreSQL services like Citus Data, and Heroku Postgres.

Open and portable

True to our commitment to openness, ParadeDB on Ubicloud is built on open source. ParadeDB itself is an open-source project. You're not locked into a proprietary system and can migrate or self-host if needed. The transparency of the Ubicloud control plane allows you to understand exactly how your service is managed and operated.

If you're interested in contributing to the underlying open-source projects, please check out the Ubicloud and ParadeDB repositories on GitHub.

What workloads are best suited for ParadeDB on Ubicloud?

Our managed ParadeDB service is ideal for a wide range of use cases, including:

  • Applications requiring both robust database functionality and advanced search capabilities.
  • Businesses looking to simplify their data stack by eliminating separate search engines.
  • Developers wanting to leverage PostgreSQL's reliability with Elasticsearch-like features.
  • Companies seeking to reduce costs and operational overhead in their search infrastructure.

Getting started

We're excited to have you try ParadeDB on Ubicloud.

To get started, provision a ParadeDB PostgreSQL database on Ubicloud here, and visit the ParadeDB docs. For your questions about search with PostgreSQL, the ParadeDB team is also available via email at and Slack.

We're committed to building an open cloud with powerful, integrated services. Your feedback and contributions are key for this mission. For any questions or comments, please reach out to us at support@ubicloud.com. We are looking forward to a better search with PostgreSQL, and always a better cloud!