I hate when people talk about disk being cheap.\u00a0 I want to just smack them.\u00a0 Disk for laptops and PC\u2019s is cheap.\u00a0 Disk for fault tolerance server computing is far from cheap.<\/p>\n
If you\u2019re running a Hyper-V cluster then you know that you can\u2019t just go out and buy some cheap storage.\u00a0 You\u2019re looking at shared storage with cluster support.\u00a0 That means either a Fibre Channel or iSCSI SAN.\u00a0 And then there\u2019s the disk.\u00a0 You could go budget and use SATA or worry about performance and go with SAS.\u00a0 Odds are it will be the latter which provides less storage for a higher price.\u00a0 Add in fault tolerance and you\u2019ve use more expensive HBA\u2019s, doubled your switch port requirements, etc.<\/p>\n
At work we\u2019re using a HP EVA with fibre channel connections via dual port HBA\u2019s and 15K disks.\u00a0 We had to worry about the ability to scale.\u00a0 We\u2019re a hosting company so unlike most virtualisation projects we had no end point.\u00a0 I couldn\u2019t say \u201cwe have 20 machines to virtualise\u201d.\u00a0 As a hosting company, 20 VM\u2019s wouldn\u2019t make much and wouldn\u2019t justify the existence of a company.\u00a0 We also had to worry about performance.<\/p>\n
When we deployed we use Windows Server 2008 Hyper-V.\u00a0 As you may know, the recommended (and VMM required) way to deploy storage for Hyper-V in a cluster is 1 LUN per VM.\u00a0 It\u2019s a storage nightmare.\u00a0 I had to be paranoid about our processes.\u00a0 LUN\u2019s, volumes, fail over cluster storage definitions and VM\u2019s all had identical names based on a naming standard I\u2019ve been using for years.\u00a0 But even with all that, I had that nightmare that I\u2019d accidentally wipe a hosted customer\u2019s VM if I deleted the wrong LUN on the SAN or someone wasn\u2019t careful with documentation.<\/p>\n
Windows Server 2008 R2 gives us the Cluster Shared Volume (CSV) for Hyper-V.\u00a0 Note the last bit; CSV is custom written by MS to be only used for Hyper-V.<\/em>\u00a0 Instead of creating lots of LUNS in the SAN and managing lots of failover cluster storage definitions and tracking documentation, you simply carve out a single LUN and deploy lots of VM\u2019s onto it.\u00a0 I\u2019ve deployed a single LUN on our new cluster.\u00a0 It\u2019s set up as a GPT volume<\/a> and used as a CSV on our W2008 R2 cluster.\u00a0 That means we can simply deploy lots of VM\u2019s onto it without worrying.\u00a0 Deleting VM\u2019s doesn\u2019t bring me to the EVA command view console to worry over accidentally deleting volumes.\u00a0 I just don\u2019t need to go there anymore unless adding more space to the CSV.\u00a0 And sure, OpsMgr will let me know when I need to do that!<\/p>\n Another nice perk of the CSV is that self service deployment on a Hyper-V cluster becomes a real world possibility.\u00a0 Sure, you could have allocated lots of individual LUNS in W2008.\u00a0 But wouldn\u2019t it be a waste if a person deployed a 50GB VM onto a 200GB LUN?<\/p>\n What about performance?\u00a0 You shouldn\u2019t suffer at all really.\u00a0 Our EVA uses a concept where all storage is striped across all disks in a disk group.\u00a0 RAID is a secondary decision when you create the LUN.\u00a0 So, when you put lots of VM\u2019s on a single CSV, they\u2019ll all use all physical disks in the CSV.<\/p>\n There\u2019s also another thing you can do.\u00a0 You can split storage for VM\u2019s across different CSV\u2019s.\u00a0 Maybe you have RAID1, RAID5 and RAID6 CSV\u2019s.\u00a0 They all are suited for different kinds of storage, even if it is virtual.\u00a0 So now your virtualised SQL could have the OS and log files on a RAID1 CSV and the data files on a RAID5 CSV.\u00a0 You get the most from the physical storage while maintaining disk performance.<\/p>\n In Windows Server 2008, dynamic disks did not meet our requirements for storage because it was too slow.\u00a0 So we went with fixed disks.\u00a0 I think our experience is not unique in any regard.\u00a0 Our customers often don\u2019t know what their storage requirements are going to be.\u00a0 We advise them to consume only what they need and take more disk later; it\u2019s a quick operation to add space to a VHD.\u00a0 But even then, physical disk gets wasted.\u00a0 Consider a Windows Server C drive.\u00a0 The recommended minimum is 40GB.\u00a0 Our average consumption is less than 50% of that.\u00a0 That means 50% of physical disk for the C: drive is wasted.\u00a0 That means a person with a 40GB C drive who is storing 13GB in the VM is really using 40GB of physical disk, plus whatever to allow for free space and save states.<\/p>\n Windows Server 2008 R2 offers greater performance for dynamic disks.\u00a0 In fact, they nearly match the performance of fixed size disks.\u00a0 We\u2019ve made the switch so we can keep costs down.\u00a0 That means that hypothetical consumer of 13GB of disk is really only consuming slightly more than that.\u00a0 Over the number of VM\u2019s I\u2019ve observed so far, we could save on 50% of our C: drives.\u00a0 Data drives are hard to figure.\u00a0 Early on we definitely save but storage requirements only ever go up.\u00a0 But we are definitely saving there too.<\/p>\n Oh yes, you can convert from a fixed VHD to a dynamic VHD.\u00a0 You just need to bring down the VM (why we won\u2019t do this to existing customers) and have space for the new VHD to be created.<\/p>\n So CSV simplifies storage administration.\u00a0 Especially if you are in an organisation where SAN management is split from server management.\u00a0 Using Dynamic Disks allows you to consume only the physical disk that you need for the data you\u2019ve stored.\u00a0 Add in other things like Live Migration, Core Parking, SLAT, improved networking, etc and you might want to do the cost benefit analysis of upgrading from Windows Server 2008 Hyper-V to Windows Server 2008 R2 Hyper-V.\u00a0 The costs of those licenses could be easily negated by the savings you\u2019ll make on storage costs (literally consume what you use) and reduced administration.<\/p>\n","protected":false},"excerpt":{"rendered":" I hate when people talk about disk being cheap.\u00a0 I want to just smack them.\u00a0 Disk for laptops and PC\u2019s is cheap.\u00a0 Disk for fault tolerance server computing is far from cheap. If you\u2019re running a Hyper-V cluster then you know that you can\u2019t just go out and buy some cheap storage.\u00a0 You\u2019re looking at … Continue reading “Keeping Hyper-V Storage Simplified and More Economic”<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_uf_show_specific_survey":0,"_uf_disable_surveys":false,"_wpcom_ai_launchpad_first_post":false,"footnotes":""},"categories":[20],"tags":[],"class_list":["post-9996","post","type-post","status-publish","format-standard","hentry","category-hyper-v"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t