I've been using Elasticsearch since early 2010. We have multiple self-hosted clusters with a few terabytes of data, used for search, analytics, and other use cases.
But recently, we started building Retrieval Augmented Generation (RAG) apps as well. For the RAG, we use Qdrant Cloud, which is good. However, I'm interested in exploring the possibility of migrating it to a self-hosted Elasticsearch setup.
Does anyone have experience using Elasticsearch as a main vector database for storage and search?
Hey Alexander, I'm happy to see you exploring our capabilities. As you start down this path, I wanted to put a few things out there for you to get a sense of the possibilities that we provide.
One, we have a general site for developers to help with the purposes of building RAG applications Elastic Search Labs
You'll be able to see tutorials, blogs, notebooks and code applications to bootstrap your journey. Perhaps of use could be taking a look at the tutorial we have here Welcome — Elastic Search Labs
Then, I'd like to point out that Elastic is doing a tremendous amount of investment into the technology that is necessary to be a fundamental part of the RAG app - so I'd probably read through a few of these blogs:
These two blogs give you a sense of our stance towards the technology that fuels our vector database capabilities:
as well as the investments we are pouring into it to make sure it's the fastest and most performant available on the market.
For RAG application developers like yourself, you may be interested in this walkthrough of RAG evaluation
And a potential look into how we built our own proprietary model that seamlessly fits into the RAG story.
If you have any questions, we'd really love to hear from you!
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.
