[LU-16757] Allow max_cached_mb to be set as a percentage of system ram Created: 21/Apr/23  Updated: 19/Sep/23  Resolved: 19/Sep/23

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: Lustre 2.15.2
Fix Version/s: None

Type: Improvement Priority: Minor
Reporter: James Beal Assignee: Patrick Farrell
Resolution: Duplicate Votes: 0
Labels: lug23dd

Issue Links:
Duplicate
duplicates LU-17030 allow setting max_cached_mb with % Resolved
Rank (Obsolete): 9223372036854775807

 Description   

When lustre clients have differing amounts of memory then setting the max_cached_mb  to an absolute number which is the same on all nodes is not likely to provide optimum results. Given vendors like to set this value and place it in the configuration log then allowing max_cached_mb to be a percentage and having the client caculate the exact value to use would be a useability win.

 

We currently do this at Sanger using a ansible templated tuning script:

 

 

{% if adjust_lustre_cache %} {% for key, value in lustre_cache.items() %}

echo We will set the lustre filesystem cache for {{ key }} to {{  (ansible_memtotal_mb * value / 100 )| int }}
lctl set_param llite.{{key*.max_cached_mb={{  (ansible_memtotal_mb * value / 100 )| int }} || true

{% endfor %} {% endif %}

 Comments   
Comment by Tommi Tervo [ 13/Sep/23 ]

Hi,

Looks like PAF fixed this:

https://jira.whamcloud.com/browse/LU-17030

Comment by Peter Jones [ 19/Sep/23 ]

Patrick do you agree that this is a duplicate of LU-17030?

Comment by Patrick Farrell [ 19/Sep/23 ]

Yes, I agree - Or perhaps the other way around, I didn't notice this one when I opened LU-17030.  But yes, LU-17030 fixes this.

Generated at Sat Feb 10 03:29:45 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.