Open Source Search Engine Battle: Solr vs Elasticsearch

We have now entered the era of a massive growth in data and cloud and this discussion is going to be pretty exciting. Applications these days generate petabytes and zettabytes of data without compromising on the speed and performance of the systems.

On top of that, when data piles up massively, searching information by steering quickly through them becomes quite a substantial back end challenge.

In this post, we shall discuss the distinct features of Solr vs Elasticsearch, the open source search engines that are gaining popularity these days.

At the end of this article, the reader will understand the different functionalities of both the search engines, and gain a fair detail insight on their individual behavior so as to decide which one to go for (as choosing one of them is not an easy task).

Origin and Building Mechanism

When it comes to Elasticsearch vs Solr, it is a fact that both were built on an open source Java library, Apache Lucene, due to which their behavior and core features are identical. Apache Lucene is a very dependable and widely deployed search engine packaged together in a group of jar files. It was first established in the year, 1991 and later in the year 2001 it became an open source project of Apache.

Apache Solr

Apache Solr, being an enterprise search platform offers search capabilities of Lucene in a very user-friendly manner. Released in the year 2008, the committers of Solr focused completely on building new search features. Gradually, distributed search mechanism became a highly desired feature.

In October 2012, SolrCloud feature was introduced, which was supposed to ease the process of distributed search. Now, Solr is high on demand and is also supported by an Apache Community comprised of 100 developers and code committers.


Elasticsearch, on the other hand, is not a product of Apache Software Foundation like Solr and Lucene. It was launched in the year 2010, just after a few years of the launch of Solr and is based at Github, a commercial software hosting service. However, it is licensed under Apache 2.0 and is an opensource distributed RESTful search engine.

The best feature of Elasticsearch is its multitenant capability, a full-text search that comes with an HTTP web interface and schema-free JSON documents.

It includes indices that can be split into shards and each one of them can have multiple replicas. Each node of Elasticsearch can contain one or more shards, and its engine also plays the role of a coordinator to assign an operation to the correct shard(s).

Elasticsearch wins this argument (Elasticsearch vs Solr) because of the major intention behind designing Elasticsearch was to fix the loopholes left in the distributed features of Solr. Hence the user might find it easier to start up an Elasticsearch cluster than that of Solr.

Major differences in features- Solr vs Elasticsearch:

Apache Solr Elasticsearch
Full-text search Distributed search
Highlighting Multi-tenancy
Faceted search Analyzer chain
Real-time indexing Analytical search
Dynamic clustering Grouping and aggregation
Database Integration  
NoSQL features and rich document handling  

 Let us look into the various grounds to understand deeply about the battle of Solr vs Elasticsearch:

1.) Coordination

Elasticsearch uses its cluster handling mechanism through an inbuilt coordination mechanism, whereas Solr uses Zookeeper. This means in order to work with SolrCloud, the user needs a Zookeeper quorum setup. People who are already using the components of Hadoop ecosystem won’t have any problem as they most likely would be having a Zookeeper quorum setup already.

2.) Shard Splitting and Rebalancing

In the chapter of Elasticsearch vs Solr, both share the Shards system feature. It is nothing but the partitioning unit for the Linux index. The user can distribute his/her index by placing the shards in a cluster on different machines. Since April 2013, Solr has supported shard splitting which allows the user to create more number of shards from the existing shards. Elasticsearch doesn’t have this feature.

However, in order to make the current system ready for sharding and addition of more machines, the user needs to have multiple shards in that machine by splitting the index based upon the estimated count of machines needed in the future.

Here, the advantage is that all the machines would be having multiple shards and when the requirement for the addition of new machines comes, Elasticsearch automatically balances the load and relocates the shards to the newly formed nodes in the cluster. The automatic shard rebalancing feature doesn’t exist in Solr.

3.) Community

Solr consists of a broad, open-source community and, hence, stands ahead in the Elasticsearch vs Solr battle. Anyone who wants to contribute to Solr can do it without any hassle, and the election of new Solr developers or code committers is held based on merit only.

Elasticsearch can be called as an open-source platform, but not completely. All its contributors have access to the source code. The users can make changes and contribute them as well. But the final changes are confirmed and done by the employees of Elastic (the company behind Elasticsearch).

This makes it clear that Elasticsearch is driven more by a single company rather than a whole community.

4.) Documentation

Solr stands unrivaled in this category. It is a perfectly documented product with clear contexts and examples of API use cases. Elasticsearch’s documentation is undoubtedly well organized, but it falls short of good examples and clear configuration instructions.

Solr vs Elasticsearch – Which is the most popular engine these days?

(This graph, clearly shows the popularity that Elasticsearch is gaining over time.)

Which one should I go for among Solr vs Elasticsearch?

Well, to be honest Elasticsearch would be an ideal pick for newer developers due to its easy-to-use nature. But, if you are already using Solr, then you better stick to it because there is no big advantage in switching to Elasticsearch.

If you are dealing with analytical queries with searching text, you better go for Elasticsearch, as Solr has search mechanism only.

Elasticsearch will always be the better choice for cloud and distributed environments that need precision in scalability and good performance. Hence, if distributed indexing is what you need, you must go for Elasticsearch.

Both the search engines have their unique features and functionalities, and now it is not hard to say which one among Solr vs Elasticsearch suits your needs.

Leave a Reply

Your email address will not be published. Required fields are marked *



About SpringPeople

Founded in 2009, SpringPeople is a global corporate training provider for high-end and emerging technologies, methodologies and products. As master partner for Pivotal / SpringSource, Elasticsearch, Typesafe, EMC, VMware, MuleSoft and Hortonworks, SpringPeople brings authentic, Certified training, designed and developed by the people who created the technology, to Corporates and Development/IT Professionals community in India. This makes SpringPeople an exclusive master certified training delivery wing, and one of the hand-picked few global partners, of these organizations - delivering their immensely popular, high-quality Certified training courses in India – for a fraction of what it costs globally.

Posts by SpringPeople