Clay Gerrard 2f0ab78f9f Shuffle disks and parts in reconstructor ... The main problem with going disk by disk is that it means all of your I/O is only on one spindle at a time and no matter how high you set concurrency it doesn't go any faster. Closes-Bug: #1491605 Change-Id: I69e4c4baee64fd2192cbf5836b0803db1cc71705		2017-01-25 18:30:17 -08:00
..
account	Make timestamp setting to be consistent in test loop	2017-01-12 17:44:10 -08:00
cli	Merge "Refactor recon to use single md5_hash_for_file function"	2017-01-24 00:38:24 +00:00
common	Merge "ISO 8601 timestamps for tempurl"	2017-01-24 22:22:04 +00:00
container	Replace 'assertTrue(a not in b)' with 'assertNotIn(a, b)'	2016-12-12 16:23:09 +07:00
obj	Shuffle disks and parts in reconstructor	2017-01-25 18:30:17 -08:00
proxy	Merge "Let users know entity size in 416 responses"	2016-12-05 19:06:45 +00:00
test_locale	monkeypatch thread for keystoneclient	2015-11-03 16:36:19 +01:00
__init__.py	Fix ZeroDivisionError in reconstructor.stats_line	2016-11-07 18:19:20 -08:00
helpers.py	Refactor tests and add tests	2016-06-15 16:36:25 +01:00