With more than three years of experience as a Database Administrator specializing in SQL Server and PostgreSQL, Adrien Obernesser focuses on performance tuning. He is skilled at diagnosing issues on customer servers and delivering actionable improvement plans, backed by solid leadership experience in cross-cultural environments. Prior to dbi services, he served as Operations Manager and Consultant DBA at Zen-Conseil. His industry experience spans the defense, telecommunications, and industrial sectors.
Introduction In the first part of this RAG series, we established the fundamentals of Naive RAG with dense vector embeddings on PostgreSQL using pgvector. That foundation works well for conceptual queries, but production systems quickly reveal a limitation: pure semantic search misses exact matches like you would have with like or Full-text searches. When someone searches for “PostgreSQL 17 performance…
Introduction Since my last series on pgvector I had quite a fun time to work on RAG workflows on pgvector and learned some valuable lessons and decided to share some of it in a blog post series on the matter. We will discover together where all RAG best practices are landing for the past 2 years and how can as…
Introduction “Make it simple.” When you have to create a solution, design an architecture, you should always take the simplest path, this is often the better solution proven by field knowledge and experiences of not scalable patterns. Which means also that you should only add the necessary and required complexity to your solutions. So while following this principle I tried…
As PostgreSQL 18 Beta 1 has been released (link) and is packed with lots of cool features like improved IO on reads for example, I wanted to test out the new pg_createsubscriber and pg_upgrade options and how they are going to help us performing smoother major upgrades using the “blue-green” migration similar to the ones available on AWS (Using Amazon…
AI projects are increasing and with it comes some risks that are or not related to this technology. Some organizations are jumping straight into “plug the LLM into a vector database” ( like Pinecone, Qdrant, pgvector…) at the risk of replaying the lift-and-shift mistakes made on the cloud projects. This time, hiding years of data-quality, governance, code, architectural debt behind…
Introduction Nowadays, everyone in IT is well versed in using ChatGPT and other LLM-powered Chat to retrieve information like Google or try to learn something with it. The problem with those models is their accuracy. Current AI models only know what information they have been fed and expose it to you in a format you understand (which is already surprisingly…
Introduction The first release of pgvector occurred in April 2021. Since then a lot has changed and is still changing. Even though we could say that the level of maturity of AI stacks and usage is currently really amazing, things are still changing fast and will probably continue to do so for a while. This is quite exciting because as…
Over the years, ORMs have industrialized query generation, allowing developers to focus on optimizing the few queries that truly need it. However, they can sometimes produce overly complex SQL with excessive joins and business logic, making tuning a nightmare. It is sometimes hard to know when you should split a query in multiple smaller parts. The complexity is amplified with…
Introduction One of the nice things with cloud providers is the low costs of their storage. Whenever you want to optimize your long-term backups and push them on Azure, you might face a wall of questions on how to do it with PostgreSQL securely and whether it would align with disaster recovery plans.The following setup is an example of how…
Introduction As PostgreSQL continues to gain traction in enterprise-grade applications, I am increasingly tasked and asked about implementing robust security policies. And, as most current DBA positions are multi-systems nowadays, I wanted to make a quick summary on roles and how they can be leveraged to implement those policies. For those accustomed to managing accounts through Active Directory Security Groups…
As I reflect on my last day at PGConf.EU 2024 in the beautiful city of Athens, I am filled with a profound sense of gratitude and excitement. The conference was a melting pot of innovation, knowledge sharing, and community spirit, all centered around PostgreSQL—a project that continues to inspire and evolve. Here are few of the talks I attended :…
Here’s the following day 2 of the PostgreSQL Conference Europe 2024 in Athens, if you haven’t read the first day here is the link. After a fantastic first day, I was excited to dive into the next day of the conference, which offered a mix of technical sessions that covered essential topics like upgrading PostgreSQL, handling statistics, and even taking…
