Find solution
How to start Apache Solr automatically?

Why Cloudera search is built on Solr and not Elasticsearch?

  • Cloudera Search uses Apache Solr for their open source search product. What made them choose Solr over Elasticsearch, which is a distributed search framework to start with. Is it incorrect to assume that the features that Cloudera has highlighted for this product can also be achieved using Elasticsearch?

  • Answer:

    Our thought process was: The feature sets of the two systems are highly similar Solr has a more mature ecosystem.  Cloudera customers could take advantage of a number of 3rd party libraries already built for Solr, not to mention a number of consulting firms who specialize in Solr We need to be able to provide support for Solr and it was readily possible to hire and grow committers to the Apache Solr project (which we have done) whereas Elastic Search doesn't have a formal model for adding outside committers The main benefit of using search in CDH vs. standalone Solr or ES is the integrated experience.  We store the indices in HDFS, integrate with MapReduce, Flume, & HBase; incorporated it into the CDH security model, manage it with Cloudera Manager etc.  A number of these enhancements were easier to add to Solr than if we had started from Elastic Search.

Charles Zedlewski at Quora Visit the source

Was this solution helpful to you?

Other answers

Solr was considered natural successor of lucene. Doug Cutting being creator of lucene and driving search effort for cloudera might have picked it up & hence it may not be a pure technical decision. elasticsearch is doing better both in terms of innovation and and simplicity (In my opinion of course) & Featurewise elasticsearch is almost on par with solr.

Praveen Manvi

Cloudera lost several large opportunities to MapR because it could not support Solr and MapR did. The start of there Solr Search integration started when they lost one very significant opportunity.  MapR supports Elastic search as well and there a several large customers using it. Ask around at Strata.

Mike Emerick

Elastic Search is relatively new player in the market. While using it you may encounter errors which are first shown or experienced by you. But the best part is that there is no half baked feature added to it. If something is there it works! Else it isn't there because, it doesn't work at all!

Ankit Bahuguna

Maybe because the Zen protocol is very faulty, and leads to all sorts of failure modes. See https://aphyr.com/posts/323-jepsen-elasticsearch-1-5-0 and compare it to http://lucidworks.com/blog/2014/12/10/call-maybe-solrcloud-jepsen-flaky-networks/

Younos Aboulnaga

Robust reliably-scalable feature-rich search on any kind of document in the repository. And as one responder says, you generally cannot be an Elasticsearch committer (and have input into the future capabilities) unless you’re an employee of Elasticsearch. I believe that is unique in open source projects I’d seen, but I can’t swear to it.They liked Solr so much, they hired it’s creator, Yonik Seeley:) Ironically, last I looked (a few months ago) Cloudera is still on Solr 4, but I suspect they’ll jump right to Solr 6… but that’s a pure WAG…

Miles Kehoe

Related Q & A:

Just Added Q & A:

Find solution

For every problem there is a solution! Proved by Solucija.

  • Got an issue and looking for advice?

  • Ask Solucija to search every corner of the Web for help.

  • Get workable solutions and helpful tips in a moment.

Just ask Solucija about an issue you face and immediately get a list of ready solutions, answers and tips from other Internet users. We always provide the most suitable and complete answer to your question at the top, along with a few good alternatives below.