Solr and LucidWorks: enterprise search for Windows Azure

Last week’s Windows Azure release delivered a host of new services for developers, ranging from hybrid cloud capabilities and Linux virtual machine support to OSS technologies delivered as a service from many vendors. Gianugo Rabellino covered the high-level view of all the exciting new offerings, and in this post I’d like to take a closer look at a service that’s likely to become very popular: LucidWorks Cloud for Windows Azure.

Lucid Imagination, the leading experts in Lucene/Solr technology, has packaged their LucidWorks Enterprise search service in a cloud-friendly way that requires only four quick and simple steps: select a plan, sign up, log in, and start using it. LucidWorks Enterprise is based on Apache Solr, the open-source search platform from the Apache Lucene project, and it includes a variety of enhancements from the search experts at Lucid that make it easy to use Lucene/Solr functionality while preserving the purity of the open source code base and open APIs. There’s a comprehensive REST API for integrating it into your applications and services, and you get all of the functionality that has made Solr and LucidWorks Enterprise so popular: high-performance indexing for a wide range of data sources, flexible searching and faceting, and user-oriented features like auto-complete, spell-checking, and click scoring.

As covered on the Lucid Imagination web site, there are four levels of service available for LucidWorks Cloud: Micro, Small, Medium and Large. Pick the level that meets your needs, sign up for the service, and you’re ready to start creating collections and searching your content. You can currently search content in web sites, Windows shares, Microsoft SharePoint sites, FTP, and other sources, with Windows Azure blob storage support coming soon. You can even index and search your data from Hadoop if desired. All index data is stored on Windows Azure drives, which offer high availability and reliability, and the Lucid dev operations engineering team can provide expert support for your LucidWorks Cloud environment.

 If you’re new to Solr, check out the free white paper available for download from the Lucid web site, which covers the basics of LucidWorks Enterprise and shows how to use the indexing and searching functionality through the LucidWorks dashboard. Most developers will want to study the API and integrate search tightly into their own software, but you can learn all of the key concepts through the dashboard UI without writing a single line of code.

One concept worth pointing out here is that Solr isn’t just about searching web sites and HTTP documents. Sure, it does a great job of that, but it can also index content stored in database tables, local file systems, and other sources. There is also an XML-based Solr document format that you can use for importing data directly into the Solr engine, giving developers flexibility for indexing any type of content from any source.

This new service from Lucid Imagination is great for those who want to get up and running quickly, but there are also developers who will want to take responsibility for all of the details and host Solr or LucidWorks Enterprise themselves. You can download LucidWorks Enterprise and install it, or you can take advantage of the simple Solr installer for Windows Azure that helps you deploy your own Solr instances as Windows Azure cloud services.

As you can see, there are many options for getting up and running with Solr and LucidWorks. For a simple overview of how easy it is to start using the new LucidWorks Cloud service, check out this Getting Started video that covers how to create a collection, index a web site, and then search that website using the Lucidworks Cloud dashboard. Lucid continues to evolve and invest in supporting the most popular Solr clients, so there will surely be more good news for Lucene/Solr users going forward.

In a future blog post, we’ll be covering how to use LucidWorks Cloud with popular content management systems such as WordPress and Drupal.