fallocate support ...

Lars Segerlund lars.segerlund at gmail.com
Wed Mar 22 04:25:18 PDT 2017


 As for the use case I am using 4 raid controllers with 12 disks on each,
and only the first 75% of the drives, ( 6 TB sas ), since the sustained
rate drops of after this, and scheduling to alternate fast and slow parts
of the disk. The difference in performance is not the issue, it's the
number of drives I have to use, 10k or 15k sas drives costs money.

 As for api, this is what I use.
 On a pristine filesystem xfs places the files allocated ( with fallocate )
from the outer to the inner tracks, which I use.
 Alternatively ext4 can make a filesystem with preallocated files through a
custom config to mkfs. This caused me to write a 'customizer' for an empty
ext4 which is offline to create the files.

 POSIX fallocate doesn't give any leeway, but perhaps the 'mode' part of
linux fallocate could be used ? ( something like hinting, outer, fasters or
similar ).

 Preferably I would like the fs to be able to rearange the disk layout to
accomodate a continous file, so a command line tool would also be fine and
perhaps preferable.

 I do see the problems involved :-D ... but when you want to store more
than a couple ot thousand T rust is sooooo much cheaper.

 / thanks , Lars.

2017-03-22 11:19 GMT+01:00 Daniel Phillips <daniel at phunq.net>:

> Hi Lars,
> On 03/21/2017 12:37 AM, Lars Segerlund wrote:
>>  Hi guys,
>>  I am doing some apps that preallocates files at known locations on disk,
>> continous and the order ( placement ) is important, so I thought I¨d ask if
>> there has been any thoughts about this on tux3 ?
>>  It¨s really a killer app for streaming, since files at the end of the
>> disk is slower, and files on the outer part of the platter is faster.
> Ah, interesting I did not realize that a factor of something less than two
> is so critical. And I guess you are telling us, spinning disks are not dead
> yet.
> Yes, this is something Hirofumi and I have discussed in the past. Tux3 is,
> in general, well suited to it. It would be required for any kind of
> real-time guarantee. We should have a specific plan for it, and see how the
> current code might be adapted for it.
>>  Preferably I¨d like to have continous files of a given size take up as
>> much of the disk as I specify, from outer cylinders to the middle. ext4 has
>> some support to set this up at filesystem creation time through a custom
>> config to mkfs.
>> So there are two issues, one is large continous file allocation, and the
>> second is file placement.
>>  Any thoughts ?
>> What API do you propose, for the user to specify these constraints?
> Regards,
> Daniel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phunq.net/pipermail/tux3/attachments/20170322/e1cc555e/attachment.html>

More information about the Tux3 mailing list