Re: btrfs hash algorithm (should xxhash be the default?)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



(I'm adding back the list)

On Tue, Feb 02, 2021 at 02:38:41PM -0500, przemek klosowski wrote:
> 
> On 2/1/21 3:34 PM, Richard W.M. Jones wrote
> >RAM disks!
> >
> >   # nbdkit memory 10G
> >   # nbd-client -b 512 localhost /dev/nbd0
> >
> Cool! I didn't know about this... the standard way I knew was via tmpfs
> 
>  mount -t tmpfs -o size=10g myrd /tmp/ramdisk
> 
> How does it compare to nbd?

tmpfs doesn't make a RAM disk block device, instead each file is
stored in memory as required.  Linux has an actual RAM disk block
device, but that's also different from nbdkit-memory-plugin.

nbdkit-memory-plugin runs in userspace and implements a sparsely
allocated RAM disk.  So you can do this on a laptop from 2021 (not
2121 or whenever it will be that we have laptops with 8 exabytes of
RAM):

  # nbdkit memory $(( 2**63 - 1 ))
  # nbd-client -b 512 localhost /dev/nbd0
  # mkfs.btrfs -K /dev/nbd0

The longer answer to this is my talk at FOSDEM:
https://archive.fosdem.org/2019/schedule/event/nbdkit/
"Better loop mounts with NBD: Take your loop mounts to the next level
with nbdkit".

> BTW, I kind-of like using loop devices, because they are persistent
> 
> dd if=/dev/zero of=/tmp/disk bs=1G count=10; mount -o loop
> /tmp/disk /mnt/disk1
> 
> Are you preferring ramdisks because of speed, or is there something else?

RAM disks are temporary by their nature.  I was suggesting this only
as a way to test the hash algos in btrfs.

nbdkit has many other plugins customized to different use cases,
including many with persistent backing; or you can write your own.

https://libguestfs.org/nbdkit.1.html

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-builder quickly builds VMs from scratch
http://libguestfs.org/virt-builder.1.html
_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx




[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [Fedora Announce]     [Fedora Users]     [Fedora Kernel]     [Fedora Testing]     [Fedora Formulas]     [Fedora PHP Devel]     [Kernel Development]     [Fedora Legacy]     [Fedora Maintainers]     [Fedora Desktop]     [PAM]     [Red Hat Development]     [Gimp]     [Yosemite News]

  Powered by Linux