SolrCluster

Solr committers Anshum Gupta and Tim Potter tell us about the upcoming Solr 5 release. We discuss making Solr “easy to start, easy to finish” while continuing to add improvements and stability for experienced users. Hear more about SolrCloud hardening, clusterstate improvements, the schema and solrconfig APIs, easier ZooKeeper management, improved flexible and schemaless indexing, and overall ease-of-use improvements.

Lucene/Solr Committers Steve Rowe and Tim Potter are back on SolrCluster to discuss how Lucidworks and the community are making changes and improvements to Solr to increase usability and add ease to the getting started experience. Steve and Tim discuss new features such as data-driven schema, start-up scripts, launching SolrCloud, and more.

This week, the SolrCluster team is joined by Rishi Easwaran from the AOL mail team. AOL has been known for their large-scale Solr deployment, with over 150B documents and 1.2B searches per day. Rishi shares AOL’s reasons for choosing Solr for mail search, an overview of their multicore architecture before migrating to SolrCloud, and their new hybrid SolrCloud architecture, its performance, scalability, cost benefits, and lessons learned throughout the process. Rishi is a principal software engineer at AOL. He has worked with the AOL mail team since 2008, as the tech lead for the existing mail search infrastructure and re-architecture efforts to implement a multi-tiered high availability, large scale hybrid SolrCloud solution. Over the years, his entire team has focused on performance, availability, and scalability of Solr. In efforts to accomplish this, they’ve delved into the source code and have developed a close working knowledge and experience with Solr. Rishi gave a more detailed talk about AOL’s Solr deployment at a recent Code Brew Meetup in Dulles. Check out the video and slides here.

Steve Rowe, Lucene/Solr Committer, PMC Member and Lucidworks Sr. Engineer, joins the SolrCluster team for a follow up discussion to the webinar he was recently featured in. Steve talks about how Solr 4 has evolved since its release in 2012, some of the major features and changes in 4.7, 4.8 and 4.9, and the trajectory of future Solr releases. Steve also gives insight into what he’s been working on for Lucidworks, side projects he’s involved in and what he’s excited about. See the “What’s New in Solr” webinar replay here. Find out more about Steve’s side project, JFlex, here. Check out the Solr Reference Guide here.

This week, we have Tim Potter, Senior Engineer at Lucidworks, Solr Committer and co-author of Solr in Action, back on the show. Tim shares with us his experience working with Solr at scale, and what led to the recent development of the Solr Scale Toolkit. The Solr Scale Toolkit is an open source project, sponsored by Lucidworks, which provides tools and guidance for deploying and managing SolrCloud in cloud-based platforms such as Amazon EC2. Read Tim’s recent blog post, Introducing the Solr Scale Toolkit.Download the Solr Scale Toolkit from GitHub here. See Tim’s slides from his recent presentation, Benchmarking Solr Performance at Scale.

On this week’s episode of SolrCluster, the team talks about Project Blacklight—an open source, front-end, discovery interface for Apache Solr—and how Blacklight is now a part of Project Hydra, a collaboration to help institutions around the world preserve, maintain and give access to their knowledge repositories and assets. Bess Sadler from the Stanford University Digital Library, a co-founder of Project Blacklight, joins us to discuss her involvement in the project, her vision for its future use and what she’s working on now. Bess Sadler is the Manager for Application Development in the Digital Library Systems group at Stanford University Library. She is one of the co-founders of Project Blacklight and Project Hydra, two open source software projects which are widely used by libraries and cultural institutions around the world. In addition to her responsibilities at Stanford, Bess is an active member in the larger library community working as a community organizer, workshop instructor, public speaker and general rabble-rouser. If you’re interested in Blacklight and Hydra, Bess also recommends the below resources. SpotlightLatest Spotlight Demo VideoGeoBlacklight

This week on SolrCluster, we are joined by Ravi Mynampaty and Katia Muser from Harvard Business School (HBS). Ravi and Katia share with us their experience migrating the Harvard Business School Alumni Website from Endeca to Lucidworks Search, Lucidworks’ commercial distribution of Solr. They explain how choosing the Solr approach saved them a ton of custom development and enabled them to offer end users features such as personalization, faceted search and navigation of various types of content (e.g., people, events, alumni stories) and autocomplete. The HBS search team chose Lucidworks Search because of the admin UI and ability to prototype very quickly. Members of the search team who did not have previous Solr experience were able to stand up, configure, and manage their Solr instance through the Lucidworks Search user interface. Katia Muser is a Technology Solution Architect at HBS.  She is a seasoned IT professional and has designed effective software solutions in various verticals such as higher education, insurance, finance, and automotive industries.  Most recently, she architected a Solr-driven search-based portal for the HBS Alumni website.  Katia has an M.S. in Mathematics and Computer Science from EPF Ecole d’Ingénieurs in France. Ravi Mynampaty refers to himself as “just another search guy standing tall on the shoulders of giants” and believes that making incremental changes is an effective path towards improving search and tries to make this happen on a daily basis.  He is currently working on Enterprise Search at HBS, which includes setting Enterprise Search strategy, design and implementation of findability standards, search tools, and search user interfaces. Ravi has masters degree in engineering from the University of Massachusetts and Brandeis University.

This week, the SolrCluster crew is joined by Lucidworks Chief of Product, Will Hayes, and Sematext Founder & President, Otis Gospodneti?. Working with Solr since its origin in 2006, Otis has a wealth of experience around trends that have come about related to search and big data technologies. Otis and Will discuss tools that Sematext has built to help monitor Solr and other stacks, manage and analyze logs, and analyze search trends, as well as the general direction that industry leaders are moving toward around data acquisition and discovery as data increasingly grows. Otis Gospodneti? is a co-author of Lucene in Action (1st and 2nd editions).  He has been involved with Lucene since 2000 and Solr since 2006 and is a member of the Lucene Project Management Committee as well as the Apache Software Foundation.  Otis is the founder of Sematext Group, Inc.  Sematext provides consulting and commercial support for Solr and ElasticSearch, and runs SPM, a Monitoring, Alerting and Anomaly Detection service, along with Search Analytics and Log Analytics services. Check out Otis’ presentation from last year’s Lucene/Solr Revolution in Dublin, Solr for Analytics: Metrics Aggregations at Sematext.

This week, the SolrCluster team is joined by Trey Grainger, Director of Engineering for Search at CareerBuilder, and Timothy Potter, Lucene/Solr Committer and Senior Engineer at Lucidworks, to discuss their recently released co-authored book, Solr in Action.  Solr in Action is a comprehensive guide to implementing scalable search with Lucene/Solr, based on the real-world applications that Tim and Trey have worked on throughout the course of their careers in Solr. Tim and Trey share with us the challenges they faced, accomplishments they achieved, and what they learned in the process of co-authoring their first book.

There’s been a lot of big news coming out of Hortonworks this week, namely announcing their partnership with Lucidworks to bring Solr-powered search to their big data platform. This partnership makes Apache Solr the defacto big data search platform of choice for all Hortonworks customers. The combination of Apache Solr and Hortonworks Data Platform (HDP) makes big data accessible to more people within an organization, not just developers, analysts, and data scientists. The SolrCluster crew was lucky enough to catch up with David McJannet, Vice President of Marketing at Hortonworks, to chat about the announcement.

This week the SolrCluster crew sits down with the Co-Founder of Lucidworks and an active member of the Lucene community – a Lucene and Solr committer, Grant Ingersoll. Grant talks about his involvement in Solr and how Lucidworks was born as well as the future of Solr and the key strengths that make Lucidworks the market leader in Enterprise Search.

In this week’s episode, SolrCluster crew chats with Wes Caldwell about the growth of big data and search in the federal space and what some of the key search/big data challenges government entities face. Wes will be presenting on this topic at Lucene/Solr Exchange DC on April 16, 2014. Wes Caldwell is the Chief Architect at Intelligent Software Solutions (ISS), a leading software company operating primarily in the public sector, headquartered in Colorado Springs, CO. He is responsible for technical oversight of a variety of programs at ISS, supporting the Intelligence Community, Department of Defense and Federal Agencies. Wes has successfully led large scale implementations of mission- critical enterprise systems deployed into strategic and tactical environments.

In this episode we discuss log analysis, data visualization, and open source with returning guest Lucidworks Chief of Product, Will Hayes, and Director of Solutions, Ravi Krishnamurthy. Listen in to hear about Lucidworks SiLK  and the new solutions and products coming out of Lucidworks this year.

In this episode we talk with the front lines of the Lucidworks Support team, Luis Lopez-Echeto and Greg Harris, about common Solr problems, how they handle complex issues, simple tips to improve your Solr instance, and the state of Solr from people who dedicate their time to helping customers get the most out of Solr.

In this episode, the SolrCluster Crew discusses how Solr actually understands and processes languages and the tricks and problems that Solr runs into when analyzing language.  We also discuss how search can introduce a new paradigm of thinking and data organization and explore a bit of the roadmap and peripheral applications Lucidworks is developing.

The SolrCluster team talks with Solr committers, Steve Rowe, Cassandra Targett, and Chris Hostetter (aka Hoss), about their involvment in the Solr community and the role of being a committer.

Our SolrCluster hosts speak with their coworker, founder, and Lucene/Solr committer, Erik Hatcher about how he found himself working on open source search and the many other projects he’s been involved in. Join us to hear from one of the people on the frontier of open source search as well as library search technology.

In this week’s show, the SolrCluster crew is talking about how document size, document count, update frequency, queries per second, query complexity, and such affect the performance of Solr and SolrCloud vs Master-Slave.

Santa’s workshop uses many different sources of information to properly manufacture and deliver toys and gifts all over the world. In this podcast, we discuss how Santa and Mrs. Claus manage their data feeds to plan their yearly distribution and keep up to date on the latest wishlist trends. Learn how the biggest little data center in the North Pole runs an efficient, elastic Solr cluster to deal with the most demanding clients of all: children.