Comments on: Performance: Thin Provisioning http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/ Building blocks for virtualization... Thu, 11 Mar 2010 09:21:29 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: Carl L http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/comment-page-1/#comment-5265 Carl L Mon, 23 Nov 2009 14:57:50 +0000 http://www.yellow-bricks.com/?p=4539#comment-5265 On the NFS question: I suspect that if you do an eager thick disk that the NFS server will have to fully allocate the space since the VMKernel will write zero's to the entire disk. Never tried it though. On the NFS question: I suspect that if you do an eager thick disk that the NFS server will have to fully allocate the space since the VMKernel will write zero’s to the entire disk. Never tried it though.

]]> By: Dennis http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/comment-page-1/#comment-5180 Dennis Wed, 18 Nov 2009 21:20:25 +0000 http://www.yellow-bricks.com/?p=4539#comment-5180 any idea if there is a big difference between vmfs version 3.31 and 3.33? Most customers will not upgrade vmfs but use the ESX 3.5 vmfs version. Many documents show that some of the metadata layout changes regarding minimized SCSI reservations were on 3.33. any idea if there is a big difference between vmfs version 3.31 and 3.33? Most customers will not upgrade vmfs but use the ESX 3.5 vmfs version. Many documents show that some of the metadata layout changes regarding minimized SCSI reservations were on 3.33.

]]> By: Duncan Epping http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/comment-page-1/#comment-5179 Duncan Epping Wed, 18 Nov 2009 19:34:18 +0000 http://www.yellow-bricks.com/?p=4539#comment-5179 Fragmentation is not an issue as usually blocks (read/write) requests from the OS are smaller than the thin chunks. Fragmentation is not an issue as usually blocks (read/write) requests from the OS are smaller than the thin chunks.

]]> By: Dennis http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/comment-page-1/#comment-5178 Dennis Wed, 18 Nov 2009 18:27:49 +0000 http://www.yellow-bricks.com/?p=4539#comment-5178 What is about fragmentation when using Thin Provisioned disks? In my opinion there must be a fragmentation of VMFS after some time especially when overallocating. What is about fragmentation when using Thin Provisioned disks? In my opinion there must be a fragmentation of VMFS after some time especially when overallocating.

]]> By: Daragh Naughton http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/comment-page-1/#comment-5174 Daragh Naughton Wed, 18 Nov 2009 15:55:20 +0000 http://www.yellow-bricks.com/?p=4539#comment-5174 A bit of digging revealed: "This behaviour of the VMware Infrastructure with a virtually provisioned NFS file system is a result of the NFS protocol, not of virtual provisioning. With NFS, storage for a Virtual Machine is not reserved in advance; it is reserved when data is actually written to the virtual machine. This is because the NFS protocol is thinly provisioned by default. Data blocks in the file system are allocated to the NFS client (ESX in this case) only when they are needed." Source: "Implementing Virtual Provisioning on EMC CLARiiON and Cellera with Virtual Infrastructure" http://www.emc.com/collateral/software/white-papers/h6131-implementing-vp-with-clariion-celerra-vmware-infrastructure-wp.pdf So it is the NFS server that only allocates the blocks - it is absolutely nothing to do with the hypervisor. Interesting. A bit of digging revealed:

“This behaviour of the VMware Infrastructure with a virtually provisioned NFS file system is a result of the NFS protocol, not of virtual provisioning. With NFS, storage for a Virtual Machine is not reserved in advance; it is reserved when data is actually written to the virtual machine. This is because the NFS protocol is thinly provisioned by default. Data blocks in the file system are allocated to the NFS client (ESX in this case) only when they are needed.”

Source: “Implementing Virtual Provisioning on EMC CLARiiON and Cellera with Virtual Infrastructure”
http://www.emc.com/collateral/software/white-papers/h6131-implementing-vp-with-clariion-celerra-vmware-infrastructure-wp.pdf

So it is the NFS server that only allocates the blocks – it is absolutely nothing to do with the hypervisor. Interesting.

]]> By: Duncan http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/comment-page-1/#comment-5156 Duncan Tue, 17 Nov 2009 14:10:19 +0000 http://www.yellow-bricks.com/?p=4539#comment-5156 No clue, somehow we must be checking a setting which is passed on to the hypervisor. But this is what people have always told me. I actually never looked into it, took it for granted. No clue, somehow we must be checking a setting which is passed on to the hypervisor. But this is what people have always told me. I actually never looked into it, took it for granted.

]]> By: Daragh Naughton http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/comment-page-1/#comment-5155 Daragh Naughton Tue, 17 Nov 2009 13:19:58 +0000 http://www.yellow-bricks.com/?p=4539#comment-5155 Hi Duncan, Regarding NFS and VMDK provisioning, the ESX Config Guide on p.99 states (as you have quoted here): "The virtual disks that you create on NFS-based datastores use a disk format dictated by the NFS server" How can the NFS Server dictate this? Surely its the ESX Server's choice, based on the type of NFS etc? Hi Duncan,

Regarding NFS and VMDK provisioning, the ESX Config Guide on p.99 states (as you have quoted here):

“The virtual disks that you create on NFS-based datastores use a disk format dictated by the NFS server”

How can the NFS Server dictate this? Surely its the ESX Server’s choice, based on the type of NFS etc?

]]> By: Eric Gray http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/comment-page-1/#comment-5099 Eric Gray Sun, 15 Nov 2009 22:53:33 +0000 http://www.yellow-bricks.com/?p=4539#comment-5099 The PDF will be listed on the VMware Technical Resource Center soon, it was just published Friday evening. I know the writer and she tipped me off early... sometimes it pays to be a VCritical subscriber. :-) Eric The PDF will be listed on the VMware Technical Resource Center soon, it was just published Friday evening. I know the writer and she tipped me off early… sometimes it pays to be a VCritical subscriber. :-)

Eric

]]> By: Duncan http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/comment-page-1/#comment-5098 Duncan Sun, 15 Nov 2009 22:42:36 +0000 http://www.yellow-bricks.com/?p=4539#comment-5098 No clue, I noticed the link on another blog... No clue, I noticed the link on another blog…

]]> By: Jason Smith http://www.yellow-bricks.com/2009/11/15/performance-thin-provisioning/comment-page-1/#comment-5092 Jason Smith Sun, 15 Nov 2009 15:58:57 +0000 http://www.yellow-bricks.com/?p=4539#comment-5092 Selecting "Support clustering features such as Fault Tolerance" when creating a VMDK will provision it eager zeroed thick. Selecting “Support clustering features such as Fault Tolerance” when creating a VMDK will provision it eager zeroed thick.

]]>