Metazeta provides tested Hadoop clusters on-demand for big data projects, experiments, teaching, and self-paced study. The Metatzeta Clusters are the result of consulting work by Paul Baclace, a software engineer with over 20 years of experience focused on infrastructure and tools. Paul started working on Nutch/Hadoop at Internet Archive with Doug Cutting in 2005 to do the DevOps needed to make index building of terabytes of data run smoothly. In 2008-2010 he helped AT&Tinteractive get up to speed with Hadoop on Amazon Web Services for a large data extraction project.
After setting up customized Hadoop clusters for a few more clients, it became clear that many companies and developers doing big data work could use pre-packaged cluster in the cloud as an alternative to Amazon's Elastic Map Reduce. The value proposition is that effort to assemble the components from different open source projects is amortized across multiple clients so the cost of doing it yourself is substantially larger than the cost of renting clusters on demand. As new releases come out, a burst of effort is needed to integrate, test, and tune the standardized cluster again. We do that for you.