The current unit tests and small-scale end-to-end tests are fine. Maybe a few could be trimmed down. Some of the long-running ones use fair-sized datasets to verify that they crunch numbers correctly.
Mahout has some algorithms for which a realistic test should take a few hours and several servers. Most of the effort in this project is aimed at these implementations. I assume that the Apache Foundation did not quite expect this level of machine support :) It would be donated by companies with an interest in Mahout. Perhaps each company could adopt one or two large-scale tests and donate server farms? If it is supposed to run huge, then it has to be tested huge. The current conversation not address this. On Mon, Dec 12, 2011 at 12:37 PM, Isabel Drost <[email protected]> wrote: > On 08.12.2011 Lance Norskog wrote: >> Given all this (especially the third), a separate Maven sub-project makes >> more sense; it should not be part of the default build. > > It may just be my personal preference, but I prefer having unit test accompany > the maven module they test. That also makes checking coverage easier. > Introducing a separate profile that is only executed on e.g. a nightly Jenkins > build as suggested by Frank earlier is useful for that. > > Isabel -- Lance Norskog [email protected]
