[lustre-devel] Using generic radix trees for large allocations
umka at cloudlinux.com
Tue Jan 26 23:21:50 PST 2021
which kernel version you use? RHEL 8.3+ or vanilla kernel 5.2+ have a lots vmalloc speedup improvements.
like separate single spin lock into two (alloc + free).
> 21 янв. 2021 г., в 21:32, James Simmons <jsimmons at casper.infradead.org> написал(а):
> One of the challenging issues for very large scale file systems is the
> performance crash when you cross about 670 stripe count. This is due to
> the memory allocations going from kmalloc to vmalloc. Once you start to
> use vmalloc to allocate the ptlrpc message buffers all the allocating
> start to serialize on a global spinlock.
> Looking for a solution the best one I found so far have been using the
> generic radix tree API. You have to allocate a page worth of data at a
> time so its cluncky for use but I think we could make it work. What do
> you think?
> lustre-devel mailing list
> lustre-devel at lists.lustre.org
More information about the lustre-devel