This is probably just a curiosity, but I was playing with ways to test the raw power of my new 8-core mac pro and was looking at how to apply this to testing.
By using multiprocessing I was able to reduce the running of the current trunk tests from 6 minutes to 3. Because a test case needs to be pickled to be pushed to the worker pool and not all test cases could be, almost 2 minutes of this is spent having to run just several tests in the main thread. If this was resolved, I think the whole test suite could be run in about a minute. When it hits a stretch of tests that can be dispatched and lights up all cores, it just screams. With the increase of multicore machines, perhaps this is something of interest. Even with fast fail - waiting for tests to run takes away from other development tasks, and being able to run the test suite more frequently would probably help catch some bugs earlier before they cascade or propagate by being extended or referred to. I'll grant this is a relatively low priority - but if anyone wants, I'd be happy to clean up my hack and post it as a gist. -Preston
-- You received this message because you are subscribed to the Google Groups "Django developers" group. To post to this group, send email to django-develop...@googlegroups.com. To unsubscribe from this group, send email to django-developers+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/django-developers?hl=en.
