Hello,
On Jan 15 18:42 Arvin Schnell wrote:
On Mon, Jan 15, 2018 at 12:35:17PM -0500, Felix Miata wrote:
Ancor Gonzalez Sosa composed on 2018-01-15 17:37 (UTC+0100):
...
For example, in the case of a typical GPT partition table in a typical
disk, the REQUIRED grain would be 1B (so we can create a partition of
any size), but the OPTIMAL grain is 1MiB (so we try to only distribute
the space in virtual "disk slices" of 1MiB)....
1MiB at the front is a convention. How did optimal *grain* get to be so big?
We do the same as parted does here. The target are SSD where the
internals are unfortunately hidden (see
https://en.wikipedia.org/wiki/Solid-state_drive#Partition_alignment,
also linked from libstorage-ng documentation):
"A typical practice for personal computers is to have each
partition aligned to start at a 1 MB (= 1,048,576 bytes) mark,
which covers all common SSD page and block size scenarios, as
it is divisible by all commonly used sizes - 1 MB, 512 KB, 128
KB, 4 KB, and 512 bytes. Modern operating system installation
software and disk tools handle this automatically."
As far as I know 1MiB would be too small for best performance
on modern flash storage, cf.
https://github.com/rear/rear/issues/1201#issuecomment-281274259
and subsequent comments and links therein.
In particular see the parted manual
https://www.gnu.org/software/parted/manual/parted.html
that reads (excerpts):
-----------------------------------------------------------
how to partition a low-end flash device
("low-end", as of 2011/2012).
For such devices, you should
use 4MiB-aligned partitions (2)
...
(2)
Cheap flash drives will be with us
for a long time to come, and, for them,
1MiB alignment is not enough.
Use at least 4MiB-aligned partitions.
For details, see Arnd Bergman's article,
http://lwn.net/Articles/428584/
and its many comments.
-----------------------------------------------------------
To make DASD same (DASD has a factor 3 in 48 KiB = 3 * 4 * 4 KiB)
a least common multiple could be 3 * 4 MiB = 12 MiB
but flash storage may have a physical block size of 8 MiB
which would make the least common multiple 3 * 8 MiB = 24 MiB
that could be used as one same default grain value
for best performance on traditional rotating disks,
on flash storage (from cheap USB sticks up to high-end SSDs),
and even on DASD.
Out ouf curiosity:
Does someone know the reason behind why IBM introduced the
unexpected factor 3 via 48 KiB required grain on DASD?
Not that I am actually surprised what IBM introduces, cf.
https://en.wikipedia.org/wiki/31-bit
Kind Regards
Johannes Meixner
--
SUSE LINUX GmbH - GF: Felix Imendoerffer, Jane Smithard,
Graham Norton - HRB 21284 (AG Nuernberg)
--
To unsubscribe, e-mail: [email protected]
To contact the owner, e-mail: [email protected]
