In most big data shops the requests for data comes from applications that are either developed in-house or are configured for custom use. Hence the data query options are many. But in several companies with active and inquisitive data scientists and analysts there will be a steady stream of ad-hoc requests. These can be handled in many different ways. Following table compares the important features of various "SQL" type interfaces to provide ad-hoc access to large data sets. Choose your technology carefully since once in place it is not easy to make a switch.