As Apache Solr becomes more powerful & easier to use, the accessibility of high quality data becomes key to unlocking the full potential of Solrs search & analytic capabilities. Traditional approaches to acquiring data frequently involve a combination of homegrown tools & scripts, often requiring significant development efforts & becoming hard to change, hard to monitor, & hard to maintain. This talk will discuss how Apache NiFi addresses the above challenges & can be used to build production-grade data pipelines for Solr. We will start by giving an introduction to the core features of NiFi, such as visual command & control, dynamic prioritization, back-pressure, & provenance. We will then look at NiFis processors for integrating with Solr, covering topics such as ingesting & extracting data, interacting with secure Solr instances, & performance tuning. We will conclude by building a live dataflow from scratch, demonstrating how to prepare data & ingest to Solr.
Bryan Bende is a Sr. Software Engineer with Hortonworks, & a PMC Member & committer on the Apache NiFi project. He has over twelve years of experience developing enterprise software solutions, & received a B.S. in Computer Science from the University of Maryland at College Park, & a M.S. in Computer Science from John Hopkins University.