Hyper-V, VMware, Xen All Compared

Techtarget has a short article that lists the pros and cons of the hardware virtualisation solutions from Microsoft, Citrix and VMware.  It’s a quick read and gives decision makers a high level comparison of the big 3 solutions.  The author does a good job of staying neutral and gives good advice.  Each solution has benefits and advantages that are unique.  Find what your real requirements are and then map those to the features.  My add-on to that: do lots of research.  Don’t take the word of a marketing or sales person.

Technorati Tags: ,,,

Poorly Planned Virtualisation Projects

This short article is an interesting read.  The author discusses how UK CIO’s feel virtualisation projects are not going to plan and aren’t returning the expected savings.  VDI is pointed out.  Yeap, VDI is one that a lot of us misunderstood early on (including me).  I’ve tried pricing it and it’s definitely more expensive than PC’s.  The management needs are at least the same if not slightly more.  But it does offer some advantages in niche areas over Terminal Services (Remote Desktop Services is too unclear now because it refers to both VDI and Terminal Services) and traditional PC’s.

Planning seems to be a problem.  That’s nothing new in IT.  The problem here is too few organisations hire the right people or bring in the right consultants.  Way too often consultants over state their skills and hiring managers hire the wrong people.  Like the old saying states: you can’t make a silk purse out of a pigs ear.

Virtualisation is complex and diverse.  It has to be thought of as a vertical foundation that becomes the bedrock for many types of IT and business application that will rest upon it.  When you have a shaky foundation, you have a shaky business.  Get it right and everything resting on it has a chance to succeed.

Technorati Tags:

VMware ESX 4.0 Update 1 Issues

I’ve just read a forum post where a VMware expert (a real one) has been reporting issues with the recent vCenter 4 and ESX 4 updates.  The latter one is scary; it can kill a host and lose some of your VM’s.  The problem happens on ESX if you manage the host using 3rd party agents.  VMware is advising that you remove the agents before an upgrade.  Another article is posted about the vCenter issue.

This is a bad play by VMware.  They’ve built up a loyal customer base but dodgy releases like this will get people interested in the possibilities of using VMM 2008 R2 to V2V migrate their VM’s onto a Hyper-V platform.

The expert in question has been advising people to stay clear of new VMware releases; let someone else test them on their own production environment.

Technorati Tags: ,,

VMM P2V Migration

Microsoft System Virtual Machine Manager (VMM) 2008 R2 includes the ability to do a P2V (physical-to-virtual) migration of Windows computers.  This is usually the last critical step in a normal virtualisation project – take those physical servers that an audit identified as being candidates to be converted into virtual machines.  The process scans the contents of the hard disk and converts them into VHD’s.  The machine specification is converted into a virtual machine configuration.

The first step in all of this begins really when you are doing a feasibility study or sizing your virtualisation hosts and storage.  You’ll run something like Microsoft’s MAP (Microsoft Assessment and Planning) toolkit.  Alternatively if you have already got Operations Manager 2007 deployed then you can install VMM 2008 R2 and wait a while before running the Virtualisation Candidates report.  That takes information from the continuous performance monitoring provided by OpsMgr.  Or you can just run individual performance reports from OpsMgr – but you need to be careful about seeing both the details and the big picture when it comes to a manual interpretation of the statistics.  And be careful about the process OpsMgr uses to store long term data.  Spikes or sudden drops may not be apparent by the data aggregation.

Once you have your Hyper-V 2008 R2 platform and VMM 2008 R2 tested, documented and in production then you can start your P2V process.

Here’s a list of the supported operating systems:

Operating System

VMM 2008

VMM 2008 R2

Microsoft Windows 2000 Server with Service Pack 4 (SP4) or later (offline P2V only)

Yes

Yes

Microsoft Windows 2000 Advanced Server SP4 or later (offline P2V only)

Yes

Yes

Windows XP Professional with Service Pack 2 (SP2) or later

Yes

Yes

Windows XP 64-Bit Edition SP2 or later

Yes

Yes

Windows Server 2003 Standard Edition (32-bit x86)

Yes (Requires SP1 or later.)

Yes (Requires SP2 or later.)

Windows Server 2003 Enterprise Edition (32-bit x86)

Yes (Requires SP1 or later.)

Yes (Requires SP2 or later.)

Windows Server 2003 Datacenter Edition (32-bit x86)

Yes (Requires SP1 or later.)

Yes (Requires SP2 or later.)

Windows Server 2003 x64 Standard Edition

Yes (Requires SP1 or later.)

Yes (Requires SP2 or later.)

Windows Server 2003 Enterprise x64 Edition

Yes (Requires SP1 or later.)

Yes (Requires SP2 or later.)

Windows Server 2003 Datacenter x64 Edition

Yes (Requires SP1 or later.)

Yes (Requires SP2 or later.)

Windows Server 2003 Web Edition

Yes

Yes

Windows Small Business Server 2003

Yes

Yes

Windows Vista with Service Pack 1 (SP1)

Yes

Yes

64-bit edition of Windows Vista with Service Pack 1 (SP1)

Yes

Yes

Windows Server 2008 Standard 32-Bit

Yes

Yes

Windows Server 2008 Enterprise 32-Bit

Yes

Yes

Windows Server 2008 Datacenter 32-Bit

Yes

Yes

64-bit edition of Windows Server 2008 Standard

Yes

Yes

64-bit edition of Windows Server 2008 Enterprise

Yes

Yes

64-bit edition of Windows Server 2008 Datacenter

Yes

Yes

Windows Web Server 2008

Yes

Yes

Windows 7

No

Yes

64-bit edition of Windows 7

No

Yes

64-bit edition of Windows Server 2008 R2 Standard

No

Yes

64-bit edition of Windows Server 2008 R2 Enterprise

No

Yes

64-bit edition of Windows Server 2008 R2 Datacenter

No

Yes

Windows Web Server 2008 R2

No

Yes

You can use the Microsoft Virtual Server 2005 Migration Toolkit (VSMT) or third-party solutions (cloning, e.g. s/w from Acronis is said to be a successful approach) for converting computers running Windows NT Server 4.0.

As if that isn’t complicate enough then you have to consider how you are going to do the P2V process.  There are two approaches:

  • Online: VMM will deploy an agent to the machine to be converted.  This is a temporary installation and does not require a license for the agent.  The agent scans the machine for suitability for an online conversion.  Upon success it will then use the Volume Shadow Copy Service (VSS) to grab file cleanly from the computer to create a new VHD for each disk in the machine.  VSS is used because things like OS files, Exchange files, SQL files, etc can be copied cleanly.  There’s a few catches with this.  (1) Not every version of (older) Windows has suitable VSS support.  VSS is a relatively new technology still.  (2) A P2V conversion is not instant.  It takes time, during which some files, particularly database files like Exchange or SQL will change after they have been copied.  That means the new VM won’t have all the data. (3) Not all server applications, e.g. MySQL or Oracle, have a VSS writer/engine.  They cannot be grabbed cleanly.  Once an online conversion is complete the source computer is left running.
  • Offline: With this process VMM deploys a boot image (Windows PE) to the machine to be converted.  The machine is reconfigured to boot from the boot image. The P2V job then runs.  The complication with this approach is that you must ensure that all the required drivers for the original physical machine must be in your boot image.  You can use the “Use storage and network drivers from the following location” option to supply additional drivers.  Because WinPE is used the physical machine must have at least 512MB RAM.

Should you use and online or offline conversion process?

Operating System on Source Computer

P2V (Online)

P2V (Offline)

Not Supported

Microsoft Windows 2000 Server Service Pack 4 (SP4)

 

X

 

The Windows Server 2003 operating systems with Service Pack 1 (SP1)

X

X

 

The Windows Server 2003 R2 Standard Edition operating system

X

X

 

The Windows XP operating systems with SP1

X

X

 

The Windows Server 2003 R2 Standard x64 Edition operating system

   

X

The Windows XP Professional x64 Edition operating system

   

X

The Windows Vista operating system

   

X

The Microsoft Windows NT Server 4.0 operating system

   

X

Again, you can use a cloning solution to work with those unsupported operating systems.

Here’s another basic rule of thumb:

  • Any machine with static data (web server) can be safely done with an online conversion.  The server stays operational and responsive to users.
  • Any machine with changing data (domain controller, Exchange, database, file, etc) should be converted with the offline approach to avoid data loss.  It does mean taking the server offline during an announced outage window.

Is there any preparation work you should do?  Yes.  Remove unwanted files.  Defrag the hard disk (schedule it in Windows Scheduled tasks, e.g. defrag C:).  Finally, remove any hardware integrated software, for example a HP server should have the HP ProLiant Support Pack from the server prior to conversion.  Failing to remove hardware integrated software will cause the new VM to blue screen or have failing services at start up.  You can do a safe mode boot and uninstall the relevant software after the P2V conversion.

When the process runs a new dynamic VHD is created by default for each physical hard disk.  You cannot reduce the size of these disks.  If you need them to be smaller then use a 3rd party solution to do this before the conversion.

When the job is complete VMM will add the Integration Components.

What about a strategy?

  • Identify virtualisation candidates.
  • Identify required drivers for offline conversions and add them to your VMM driver pool.
  • Prepare the physical computer, e.g. do a defrag and double check anti virus, etc.
  • Make sure all backups of the physical computer have worked OK and that you can recover from any disaster.
  • Maybe do an online conversion to test the process for the server in question.  Place the new VM on a test virtual network.  Make sure it boots up OK and performs OK.  This won’t affect the production physical server.
  • Perform final P2V preparations, e.g. uninstall hardware integrated software.
  • Perform a suitable conversion (probably offline)of the physical computer.  Leave it offline.  Bring the VM online and test it.
  • Put the new VM into production.
  • Make sure backups are working OK for the new VM.
  • Leave the physical server powered off for a pre-agreed timeframe before removing/recycling the physical computer.  You never know what will happen, e.g. require a
  • reversal of the process to V2P (not in VMM) of the server.

Notes:

  • FAT/FAT32 cannot be converted using Online P2V
  • You can do a P2V of virtual machines, however VMware users will want to use the V2V approach.
  • You cannot do a P2V of an in-place cluster.  However you can convert each cluster node and then create a new failover cluster.

Technorati Tags: ,,

Virtualisation: The Undersold Truth

Ease of administration.

To a sys admin, those 3 words mean a lot.  To a decision maker like a CIO or a CFO (often one and the same) they mean nothing.

It’s rare enough that I find myself working with physical boxes these days.  Most everyone is looking for a virtualised service which is cool with me.  Over the last 2 weeks I’ve been doing some physical server builds with Windows Server 2008 R2.  I know the techniques for a automated installation.  I just haven’t had time to deploy them for the few builds I needed to do.  Things like Offline Servicing for VM’s and MDT/WDS (upgrade) are in my plans but things had to be prioritised.  I’ve just kicked off a reboot of a blade server.  By the time that’s finished it’s POST I’ll have made and half drunk a cup of coffee.  After working with VM’s almost exclusively for the last 18 months, working with a physical box seems slow.  These are fine machines but the setup time required seems slow.  Those reboots take forever!  VM reboots: well there’s no POST and they reboot extremely quickly.

Let’s compare the process of deploying a VM and a physical box

Deploy a VM

  • Deploy a VM.
  • Log in and tweak.
  • Handover the VM.

Notes on this:

  • The free Offline Servicing Tool can allow you to deploy VM’s that already have all the security updates.
  • This process can be done by a delegate “end user” using the VMM self servicing web interface.
  • The process was probably just an hour or two from end to end.

Deploy a Physical Server

  • Create a purchase request for a new server.
  • Wait 1-7 days for a PO number.
  • Order the server.
  • Wait for up to 7 days for the server to be delivered.
  • Rack, power and network the server.
  • We’ll assume you have all your ducks in a row here: Use MDT 2010 or ConfigMgr to deploy an operating system.
  • The OS installs and the task sequence deploys updates (reboots), then applications (reboots), then more updates (reboots) and then makes tweaks (more updates and a reboot).
  • You have over the server.

Notes on this:

  • Most people don’t automate a server build.  Manual installs typically take 1 to 1.5 days.
  • There will probably be up to 1 day of a delay for networking.
  • The “end user” can’t do self service and must wait for IT, often getting frustrated.
  • The entire process will probably take 10.5 to 16.5 days.

Total Hardware Breakdown

Let’s assume the VM scenario used a cluster.  If the hardware failure crashed the host then the VM stops running.  The cluster moves the VM resource to another host (VMM will choose the most suitable one) and the VM starts up again.  Every VM on the cluster has hardware fault tolerance.  If the hardware failure was non-critical then you can use Live Migration to move all the VM’s to another host (VMM 2008 R2 maintenance mode) and then power down the host to work on it.  There’s no manual intervention at all in keeping things running.

What if you used standalone (un-clustered) hosts.  As long as you have an identical server chassis available you can swap the disks and network cables to get back up and running in a matter of minutes.

Unbelievably worst case scenario with un-clustered hosts: you can take the data disks and slap them into another machine and do some manual work to get running again.  As long as the processor is from the same manufacturer you’re good to go in a few hours.

If a physical box dies then you can do something similar to that.  However, physical boxes tend to vary quite a lot.  A farm of virtualisation hosts don’t usually vary too much at all.  If a DL380 dies then you can expect to put the disks into a DL160 and have a good result.  It might work. 

Most companies don’t purchase the “within 4 hours” response contracts.  And even if they do, some manufacturers will do their very best to avoid sending anyone out by asking for one diagnostic test after another and endless collections of logs.  It could be 1 to 3 days (and some angry phone calls) before an engineer comes out to fix the server.  In that time the hosted application has been offline, negatively affecting the business and potentially your customers.  If only a physical server was a portable container like a VM – see boot from VHD.

Summary

You’ve heard all those sales lines on virtualisation: carbon footprint, reduced rack space, lower power bills, etc.  Now you can see how easier administration can make your life easier but positively impact the business.

My experience has been that when you translate techie-speak into Euros, Dollars, Pounds, Rubles, Yen or Yuan then that get’s the budget owners attention.  The CFO will sit up and listen and probably decide in your favour.  And if you can explain how these technologies will have real positive impacts on the business then the other decision makers will also have your attention.