[LU-3970] Add procfs interface for clearing lustre caches in parallel Created: 18/Sep/13 Updated: 09/Feb/16 Resolved: 18/Jul/14 |
|
| Status: | Resolved |
| Project: | Lustre |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Improvement | Priority: | Minor |
| Reporter: | Ryan Haasken | Assignee: | Jian Yu |
| Resolution: | Won't Fix | Votes: | 0 |
| Labels: | patch, performance | ||
| Issue Links: |
|
||||||||
| Epic/Theme: | Performance | ||||||||
| Rank (Obsolete): | 10579 | ||||||||
| Description |
|
Cray experienced slowness clearing the lustre caches after job termination. This slowness is a result of clearing the caches for each namespace in series. Lustre should provide a high-level procfs interface which clears lustre caches across namespaces in parallel and returns when all caches are cleared. The interface should be at /proc/fs/lustre/ldlm/drop_caches, and when written to, it must clear all lustre caches in parallel and return when finished. It must have the same effect as the following, but in parallel: for LRU in $(ls /proc/fs/lustre/ldlm/namespaces/*osc*/lru_size); do
echo clear > $LRU
done
for LRU in $(ls /proc/fs/lustre/ldlm/namespaces/*mdc*/lru_size); do
echo clear > $LRU
done
|
| Comments |
| Comment by Ryan Haasken [ 18/Sep/13 ] |
|
I am working on uploading a patch to Gerrit. |
| Comment by Oleg Drokin [ 20/Sep/13 ] |
|
I wonder if why doing hte echos in parallel won't work? |
| Comment by Ryan Haasken [ 20/Sep/13 ] |
|
That would work as well. However, Cray's ALPS (Application Level Placement Scheduler) team requested that Lustre provide a higher level interface which clears all the caches in parallel. Do you feel that this is an appropriate enhancement? I've already tested a patch, but I am still learning how to submit it to Gerrit for review. |
| Comment by Ryan Haasken [ 26/Sep/13 ] |
|
I have submitted a patch to Gerrit: http://review.whamcloud.com/#/c/7783/ |
| Comment by Ryan Haasken [ 02/Jun/14 ] |
|
Can somebody please mark this bug related to
|
| Comment by Cory Spitz [ 18/Jul/14 ] |
|
http://review.whamcloud.com/#/c/7783 is abandoned in favor of the approach taken with This bug ought to be closed now. Jian, can you please make it so? |
| Comment by Peter Jones [ 18/Jul/14 ] |
|
ok Cory |