https://jira.whamcloud.com This file is an XML representation of an issue en-us 9.4.14 940014 05-12-2023 https://jira.whamcloud.com/browse/LU-10967 Lustre <p>Opening this as somewhat of a catch all for improvements to MDT memory management, either through pinning or making more selective use of the page cache.  This will be important for at least two cases...<br/> A) Data On MDT<br/> B) Combining OST and MDT on a single server<br/> MDT performance historically relied on "read mostly" workload to backing storage, with much of the important and frequently used metatdata in DRAM cache.  With increase in data blocks from either case above, that could quickly get pushed out of cache and thereby slow down metadata ops.</p> <p>It may not be practical to try to pin all MDT "real metadata" info in memory, but some features to perhaps release DoM from the page cache faster, or refresh the MDT (and OST) bitmaps and inode tables, would be helpful.</p> Relevant to any configuration where critical MDT metadata may not fit in memory or may get pushed out of DRAM cache due to other IO. LU-10967

MDT page cache management improvements

Improvement Minor Open Unresolved Peter Jones Nathan Dauchy Mon, 30 Apr 2018 13:28:48 +0000 Mon, 30 Apr 2018 21:20:14 +0000 Lustre 2.10.3 0 8 <p>The other tickets I could find that are even marginally related to this issue...<br/> <a href="https://jira.whamcloud.com/browse/LU-1885" title="Lustre should be able to limit amount of memory used for read and write caches" class="issue-link" data-issue-key="LU-1885"><del>LU-1885</del></a> : Lustre should be able to limit amount of memory used for read and write caches<br/> <a href="https://jira.whamcloud.com/browse/LU-2477" title="Poor MDS create performance due to ARC cache growth" class="issue-link" data-issue-key="LU-2477"><del>LU-2477</del></a> : Poor MDS create performance due to ARC cache growth<br/> <a href="https://jira.whamcloud.com/browse/LU-10946" title="add an interface to load ldiskfs block bitmaps" class="issue-link" data-issue-key="LU-10946"><del>LU-10946</del></a> : add an interface to load ldiskfs block bitmaps</p> <p>Nathan</p> <p>Thanks for opening this ticket. It will be good to get a variety of views onto possible approaches here.</p> <p>Peter</p> <p>The proposal for <a href="https://jira.whamcloud.com/browse/LU-10946" title="add an interface to load ldiskfs block bitmaps" class="issue-link" data-issue-key="LU-10946"><del>LU-10946</del></a> to pin the block and inode bitmaps would also help for the MDS, though to a lesser extent because the MDT is typically flash so reading them on demand from storage will be much faster than competing with a streaming I/O workload on the OST on spinning disks.</p> <p>The block bitmaps would need 4KB of RAM per 128MB of MDT space, and the inode bitmaps would double that. For a largish MDTs, say 8TB (=4B files), this would mean 4KB * (8TB/128MB) = 256MB for the block bitmaps, and an additional 256MB for the inode bitmaps, which is not unreasonable. Pinning the inode <b>tables</b> on the MDT would be totally impractical, since they consume (by default) 1/2 of all the MDT space (i.e. you would need 4TB of RAM to pin all of the MDT inode tables).</p> Related LU-15 LU-3631 LU-1885 LU-10946 Development Rank 1|hzzwiv: Rank (Obsolete) 9223372036854775807