[sheepdog] [PATCH v2] sheep/md: change process of for_each_object_in_stale() to multi-threads
namei.unix at gmail.com
Wed Feb 26 04:57:56 CET 2014
On Wed, Feb 26, 2014 at 11:43:28AM +0800, Robin Dong wrote:
> In our test environment, we upload 6TB data and kill one node, then the sheep
> daemons on each server will try to rename files from 'data' to '.stale', but
> the rename operations cost almost half an hour. And in this half an hour, the
> whole cluster can't be read or write.
> To accelerate the speed of renames, we change for_each_object_in_stale() to
> multi-threads, which is one thread for one disk.
I think you are optimizing the wrong place. You should multi-thread
instead of for_each_object_in_stale
More information about the sheepdog