Feature #2421
closed
crush: quantitatively validate mapping quality
Added by Sage Weil almost 12 years ago.
Updated almost 7 years ago.
Description
- measure variance of resulting mapping vs expected (based on weights)
- give some quantitative comparison with expected variance
- include tests when some devices are marked out or partially out
- for random/uniform subset of buckets, and those buckets' items
(so we can e.g. test small subset of buckets with most/all devices out)
- Description updated (diff)
- Category set to 10
- Translation missing: en.field_story_points set to 8
- Translation missing: en.field_position deleted (
6)
- Translation missing: en.field_position set to 6
- Target version set to v0.48
- Translation missing: en.field_position deleted (
6)
- Translation missing: en.field_position set to 1251
- Assignee set to caleb miles
- Start date set to 05/17/2012
- % Done changed from 0 to 30
- Status changed from New to Resolved
- Translation missing: en.field_position deleted (
1259)
- Translation missing: en.field_position set to 1259
- Project changed from Ceph to RADOS
- Category deleted (
10)
- Target version deleted (
v0.48)
Also available in: Atom
PDF