Drill will be used by analysts and developers who are doing interactive analysis of large-scale datasets. It is intended for ad hoc, fast query where there are multiple data sources and formats. Previously that required writing Java programs, which is neither ad hoc nor fast. Drill plans to change that.
Inspired by Google Dremel, Drill fills the gap between batch processing (Hadoop MapReduce) and stream processing (Storm and Apache S4). It will provide flexibility for data sources, using structured and unstructured data. The flexibility and speed of Drill will make things possible that otherwise have not been feasible.
So, how will you use Drill? What use cases? What data sources? We’d like to hear from you. You can add your comments and suggestions on the Apache Drill mailing lists.
Of course, the community welcomes new and disparate voices. Anyone can get involved with the Apache Drill project, as it is the only truly open source project dedicated meeting these needs.
Join the mailing list and learn where your skills can be brought to bear. You can also check out the source code by following the directions on the Drill project web site. Follow Drill news via @ApacheDrill.