mongo-java-driver VS pgvector

To support the exploration, I've developed a simple Retrieval Augmented Generation (RAG) workflow that works completely locally on the laptop for free. If you're interested, you can find the code itself here. Basically, I've used Testcontainers to create a Postgres database container with the pgvector extension to store text embeddings and an open source LLM with which I send requests to: Meta's llama3 through ollama.

# postgres image with `pgvector` enabled FROM postgres:16.3 RUN apt-get update \ && apt-get install -y postgresql-server-dev-all build-essential \ && apt-get install -y git \ && git clone https://github.com/pgvector/pgvector.git \ && cd pgvector \ && make \ && make install \ && apt-get remove -y git build-essential \ && apt-get autoremove -y \ && rm -rf /var/lib/apt/lists/* EXPOSE 5432

I helped work on the RAG part of this :-)

We used https://github.com/pgvector/pgvector under the hood and found it extremely easy to integrate with our database schema - being able to just specify the structure of a table and have metadata fields alongside the embeddings made the code very easy to reason about.

# Install Postgres and pgvector !apt-get update && apt install postgresql postgresql-server-dev-14 !git clone --branch v0.6.2 https://github.com/pgvector/pgvector.git !cd pgvector && make && make install # Start database !service postgresql start !sudo -u postgres psql -U postgres -c "ALTER USER postgres PASSWORD 'pass';"

When talking about Vector Databases, in the market we can find the specialized ones and multi-model, most of the major database providers like Oracle, PostgreSQL or MongoDB, for mention some of them, have integrated a specific solution to retrieve vector data.

For this example, I wanted to focus on how pgvector  – an open-source vector similarity search for Postgres – can be used to identify data similarities that exist in enterprise data.

pgvector supports dense vector search well, but it does not have plan to support sparse vector.

There's an issue in the pgvector repo about someone having several ~10-20million row tables and getting acceptable performance with the right hardware and some performance tuning: https://github.com/pgvector/pgvector/issues/455

I'm in the early stages of evaluating pgvector myself. but having used pinecone I currently am liking pgvector better because of it being open source. The indexing algorithm is clear, one can understand and modify the parameters. Furthermore the database is postgresql, not a proprietary document store. When the other data in the problem is stored relationally, it is very convenient to have the vectors stored like this as well. And postgresql has good observability and metrics. I think when it comes to flexibility for specialized applications, pgvector seems like the clear winner. But I can definitely see pinecone's appeal if vector search is not a core component of the problem/business, as it is very easy to use and scales very easily