I put a real search engine into a Lambda, so you only pay when you search

Great to see more love for using Lucene on Lambda via GraalVM native binaries! Explored that same idea few years back for my personal blog [1]. It worked pretty well, but I moved eventually away from Lambda as I wanted to close down my personal AWS account. One thing I always meant to explore was embedding the (in my case small) Lucene index directly into the binary image using Lucene's RAMDirectory via Quarkus' bytecode recording machinery, but I never got to it.

[1] https://www.morling.dev/blog/how-i-built-a-serverless-search...

This is great!

I have been using Quickwit as a low cost search engine on Lambda. It works very well for my relatively small and infrequently updated dataset.

Unfortunately Quickwit devs have decided to not support the Lambda deployment mode going forward so eventually I'll need another option.

Fascinating. In a modern app I find that search quickly can become one of the most expensive aspects. It’s not trivial to implement on your own and with the growing effort on hybrid style search, the costs continue to grow. I love work like this.

Maybe a stupid question: why can't the indexer write directly to EFS, why does it write to s3 and lambda downloads it to EFS?

Serverless gets a lot of hate for its name, but I have had so much success with it myself that I have moved on to its successor and have started developing for computerless architectures.

"128MB default with up to 3008MB max. You can submit a support ticket to get 10GB RAM, but I was too lazy to argue with AWS support."

Was this written before wide availability of 10g memory lambdas?