Comments on: Multiple VMFS volumes per LUN = Not a Good Idea http://www.punchingclouds.com/?p=1020 Breakthroughs in the Field of Virtualization, Cloud Computing and other Cool Shit! Tue, 13 Jul 2010 07:23:49 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: Bill Griffith http://www.punchingclouds.com/?p=1020&cpage=1#comment-420 Bill Griffith Mon, 08 Mar 2010 17:31:42 +0000 http://www.punchingclouds.com/?p=1020#comment-420 Derek, why not just do a volume grow? Then you don't have to worry about metadata, locks, multiple luns etc... Derek, why not just do a volume grow? Then you don’t have to worry about metadata, locks, multiple luns etc…

]]> By: Bill Griffith http://www.punchingclouds.com/?p=1020&cpage=1#comment-419 Bill Griffith Mon, 08 Mar 2010 17:05:26 +0000 http://www.punchingclouds.com/?p=1020#comment-419 This doc doesn't explicitly talk about the locking mechanism (where is that doc???) it does talk about performance of thin vs thick provisioned disks. www.vmware.com/pdf/vsp_4_thinprov_perf.pdf This doc doesn’t explicitly talk about the locking mechanism (where is that doc???) it does talk about performance of thin vs thick provisioned disks.

http://www.vmware.com/pdf/vsp_4_thinprov_perf.pdf

]]> By: Coral Balla http://www.punchingclouds.com/?p=1020&cpage=1#comment-405 Coral Balla Fri, 05 Mar 2010 09:41:15 +0000 http://www.punchingclouds.com/?p=1020#comment-405 I like the layout of your blog and I'm going to do the same thing for mine. Do you have any tips? Please PM ME on yahoo @ AmandaLovesYou702 I like the layout of your blog and I’m going to do the same thing for mine. Do you have any tips? Please PM ME on yahoo @ AmandaLovesYou702

]]> By: Derek http://www.punchingclouds.com/?p=1020&cpage=1#comment-390 Derek Mon, 01 Mar 2010 18:36:30 +0000 http://www.punchingclouds.com/?p=1020#comment-390 That makes sense as it relates to multiple datastores, but is this really relevant if the multiple VMFS partitions are different extents of the same VMFS datastore? An example I have had is a LUN on an iSCSI enclosure is created, a VMFS file system and partition on the LUN. Later we needed more space so we extended the LUN, then added a VMFS extent to the LUN. I could have added another LUN with the free space and then added an extent there, but I have had issues on occasion with other storage admins deleting LUN's they think are not in used, leaving me with a damaged VMFS file system. That is another story. In any case, is there any issue you could think of with 2 extents of the same VMFS on one extended LUN? The fact that both Extents locked at the same time shouldn't be any more issue than if it was one extent. Either way all the space would be locked. Just curious to your opinion. That makes sense as it relates to multiple datastores, but is this really relevant if the multiple VMFS partitions are different extents of the same VMFS datastore?

An example I have had is a LUN on an iSCSI enclosure is created, a VMFS file system and partition on the LUN. Later we needed more space so we extended the LUN, then added a VMFS extent to the LUN. I could have added another LUN with the free space and then added an extent there, but I have had issues on occasion with other storage admins deleting LUN’s they think are not in used, leaving me with a damaged VMFS file system. That is another story. In any case, is there any issue you could think of with 2 extents of the same VMFS on one extended LUN? The fact that both Extents locked at the same time shouldn’t be any more issue than if it was one extent. Either way all the space would be locked.

Just curious to your opinion.

]]>