At work, we do tech-talks every few weeks and this week I was asked to do a tech-talk on Elasticsearch. I have been learning and using Elasticsearch to implement a 60 million document search solution over the last few months. It has worked really well and has been a joy to use. It’s been fun to work in a different domain for awhile and get a handle on the world of search.
Elasticsearch is an impressive search server packed full of features. I initially evaluated a few of the search solutions out there, most notably Apache Solr, and decided upon Elasticsearch because of its ease of use and rock solid distributed architecture. I love the REST JSON API, schema-less indexing and full document source retrieval ability.
As I’ve implemented Elasticsearch to support a new feature we are rolling out, I’ve already thought about other areas of our application that could take advantage of a search index; to take load off of our RDBMS and provide a performance boost to our stack. Also, I see Elasticsearch as a fantastic way to handle Business Intelligence and Big Data concerns with tools like Kibana sitting on top of it.
Here is my slide deck from the tech-talk this week: