Server version details - Lustre 188.8.131.52
Client version details - Lustre 2.11.0-1 and 2.7.19 (old version) tested on both
We have a Lustre client running a primarily compute intensive operation, but which acts over a medium sized data set (~ 80 GB) consisting of a number of files that are ~ 300-500MB big. We're trawling across those files sequentially in a repeated fashion.
We'd like to retain the data set within the Lustre client cache, and have increased this using lctl set_param llite.*.max_cached_mb. However, one observation we've made is that the cache usage appears to top out at ~ 93%, i.e. when configured at the default ~ 70304 MB for this host (140GB mem), the used_mb appears to top out at 65920:
Likewise, when we up the max to 100000MB, it seems to top out at ~93GB ish (as in we see the used_mb fix, and an increase in network traffic on the Lustre port back to the lustre server).
I'm wondering whether that is expected, and whether there's any configuration option possible to max used of that remaining portion of the cache?
15 St. Botolph Street
London, EC3A 7DT
Phone: +44 (0)20 7663 3754
This email is intended solely for the addressee and is strictly confidential. If you are not the intended addressee, please email it back to the sender and then immediately delete it. Any unauthorised dissemination, distribution, copying or other use of this message is strictly prohibited. The information contained in this email and any attachments are for information purposes only. Whilst virus protection checking is in place, we accept no liability in respect of non-detected viruses.
Tyler Capital Partners LLP is registered in England with company number OC368810 and is Authorised and Regulated by the Financial Conduct Authority. Tyler Capital Ltd is registered in England with company number 04971692. VAT number 927124041.
Registered address: 15 St. Botolph Street, London EC3A 7DT