<html><body><p>GPFS will issue 4K IOs if the file system is 4K-aligned, which is the default for 4.1.1+, and may issue requests as small as 512 bytes otherwise. If a given array presents 4K physical block size to the world, you won't be able to format a non-4K-aligned file system which puts metadata on that disk. It is generally believed that 4K IO size produces optimal performance with flash-based storage in general, because this is the native physical block size, and 512-byte sector emulation entails doing a read-modify-write. We don't have the data on how much difference the sector size actually makes. If in doubt, I would recommend going with 4K sectors and a 4K-aligned file system.<br><br>yuri<br><br><img width="16" height="16" src="cid:1__=07BBF52EDFF9D1FB8f9e8a93df938690918c07B@" border="0" alt="Inactive hide details for "Simon Thompson (Research Computing - IT Services)" ---05/16/2016 02:48:32 AM---Hi Paul, Thanks, inte"><font color="#424282">"Simon Thompson (Research Computing - IT Services)" ---05/16/2016 02:48:32 AM---Hi Paul, Thanks, interestingly the FS-900 only does raid5. I get no choice about that, I do get a se</font><br><br><font size="2" color="#5F5F5F">From: </font><font size="2">"Simon Thompson (Research Computing - IT Services)" <S.J.Thompson@bham.ac.uk></font><br><font size="2" color="#5F5F5F">To: </font><font size="2">"'gpfsug-discuss@spectrumscale.org'" <gpfsug-discuss@spectrumscale.org>, </font><br><font size="2" color="#5F5F5F">Date: </font><font size="2">05/16/2016 02:48 AM</font><br><font size="2" color="#5F5F5F">Subject: </font><font size="2">Re: [gpfsug-discuss] Flash for metadata</font><br><font size="2" color="#5F5F5F">Sent by: </font><font size="2">gpfsug-discuss-bounces@spectrumscale.org</font><br><hr width="100%" size="2" align="left" noshade style="color:#8091A5; "><br><br><br><font size="4"><br>Hi Paul,<br><br>Thanks, interestingly the FS-900 only does raid5. I get no choice about that, I do get a sector size choice when I create a lun on it...<br><br>I was also wondering how gpfs would do a metadata update. Would it be smart enough to know the sector size was 512b and only modify that, or would it always do a read modify write on the whole 4k inode?<br><br>Simon<br><br>-----Original Message-----</font><b><font size="4"><br>From: </font></b><font size="4">Sanchez, Paul [</font><a href="mailto:Paul.Sanchez@deshaw.com"><u><font size="4" color="#0000FF">Paul.Sanchez@deshaw.com</font></u></a><font size="4">]</font><b><font size="4"><br>Sent: </font></b><font size="4">Monday, May 16, 2016 10:28 AM GMT Standard Time</font><b><font size="4"><br>To: </font></b><font size="4">gpfsug-discuss@spectrumscale.org</font><b><font size="4"><br>Subject: </font></b><font size="4">Re: [gpfsug-discuss] Flash for metadata<br></font><br><font color="#1F497D" face="Calibri">Simon,</font><br><font color="#1F497D" face="Calibri"> </font><br><font color="#1F497D" face="Calibri">I can’t speak specifically to the FS-900, since I generally use Dell Compellent and NetApp EF arrays for flash today. But I also take care to ensure that the system pool metadata NSDs use RAID10 (or equivalent) rather than RAID5/6. For small 4K metadata writes, on a 4K blocksize LUN this could result in optimal writes. For larger native blocksize LUNs, it eliminates the parity calculation for read-modify-write operations. </font><br><font color="#1F497D" face="Calibri"> </font><br><font color="#1F497D" face="Calibri">Thanks,<br>Paul</font><br><font color="#1F497D" face="Calibri"> </font><br><b><font face="Tahoma">From:</font></b><font size="4" face="Times New Roman"> </font><font face="Tahoma">gpfsug-discuss-bounces@spectrumscale.org [</font><font face="Tahoma"><a href="mailto:gpfsug-discuss-bounces@spectrumscale.org">mailto:gpfsug-discuss-bounces@spectrumscale.org</a></font><font face="Tahoma">] </font><b><font face="Tahoma">On Behalf Of</font></b><font face="Tahoma"> Simon Thompson (Research Computing - IT Services)</font><b><font face="Tahoma"><br>Sent:</font></b><font face="Tahoma"> Monday, May 16, 2016 9:44 AM</font><b><font face="Tahoma"><br>To:</font></b><font face="Tahoma"> gpfsug-discuss@spectrumscale.org</font><b><font face="Tahoma"><br>Subject:</font></b><font face="Tahoma"> [gpfsug-discuss] Flash for metadata</font><br><font size="4" face="Times New Roman"> </font><br><font face="Tahoma">Hi all</font><br><font face="Tahoma"> </font><br><font face="Tahoma">I'm currently about to deploy some flash systems (FS-900) to take our metadata for one of our Spectrum Scale file-systems. We use 4k inodes for this file-system.</font><br><font face="Tahoma"> </font><br><font face="Tahoma">On the FS-900, when creating a volume, we get a choice of 512b or 4k sectors. Does anyone have a suggestion on this? On a spinning disk RAID array, I get what a strip size is and we'd use multiples of the stripe where possible for the block size.</font><br><font face="Tahoma"> </font><br><font face="Tahoma">Is the sector size on the FS-900 the smallest chunk that can be written out? And so would it make sense to use 4k sectors for 4k GPFS inodes right?</font><br><font face="Tahoma"> </font><br><font face="Tahoma">Related. In the deployment guide for the FS-900 arrays, if one is using a file-system (e.g. XFS) on the array the guide talks about aligning the partitions with with array, when GPFS is used on a storage array, does it automatically work out the sector alignment?</font><br><font face="Tahoma"> </font><br><font face="Tahoma"> </font><br><font face="Tahoma">Thanks</font><br><font face="Tahoma"> </font><br><font face="Tahoma">Simon</font><tt>_______________________________________________<br>gpfsug-discuss mailing list<br>gpfsug-discuss at spectrumscale.org<br></tt><tt><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss">http://gpfsug.org/mailman/listinfo/gpfsug-discuss</a></tt><tt><br></tt><br><BR>
</body></html>