Today’s post is from Kathy Davies one of the excellent Hyper-V technical writers. Enjoy!
Here is some preliminary documentation on planning for backing up Hyper-V VMs. It will eventually be made part of our Planning and Deployment guide on TechNet.
When you plan a backup and recovery strategy for a virtualized server environment, there are several factors to consider. You must consider the different types of backups you can make, the state of the virtual machine, and the type of storage being used by the virtual machines. This topic discusses the advantages, disadvantages, and considerations for these factors.
The backup integration service and the Hyper-V Volume Shadow Copy Service (VSS) Writer provide the mechanism for backing virtual machines as well as system-wide settings that apply to Hyper-V. There are two basic methods you can use to perform a backup. You can:
· Perform a backup from the server running Hyper-V. Using this method to perform a full server backup is the recommended method because it captures more data than the other method. If the storage is compatible with Hyper-V and the Hyper-V VSS Writer, you can perform a full server backup that helps protect all of the data required to fully restore the server. The data included in such a backup includes the configuration of virtual machines and virtual networks, snapshots associated with the virtual machines, and virtual hard disks used by the virtual machines. As a result, using this method can make it easier to recover the server if the need arises, because you will not have to recreate virtual machines or system-wide settings such as virtual networks.
· Perform a backup from within the guest operating system of a virtual machine. This method is useful when you need to back up data from storage that is not supported by the Hyper-V VSS writer.
As you plan your backup strategy, consider the compatibility between the storage and backup solutions:
· Virtual hard disks offer the best compatibility and can be stored on many types of physical media. For more information about the types of storage you can use with Hyper-V, see “Hardware Considerations” in the Hyper-V Planning and Deployment Guide on TechNet (http://technet.microsoft.com/en-us/library/cc816844.aspx).
· Network-based storage such as shared folders should be used with caution. If the network-based storage is unavailable when a backup is attempted, the backup will fail.
· Physical disks that are directly attached to a virtual machine (sometimes referred to as ‘pass-through disks’) cannot be backed up by the Hyper-V VSS writer. As a result, this type of disk will not be included in any backup performed by a backup program that uses the Hyper-V VSS writer. In this situation, you would need to use some other process to back up the physical disk. For example, you could run a backup of the data on the iSCSI storage from a backup application running in the guest operating system.
· Storage accessed from a virtual machine by using an Internet SCSI (iSCSI) initiator within the guest operating system will not be included in a backup of the physical computer. In this scenario, you must use another process to back up the data from the iSCSI-based storage before you perform a full server backup. For example, you could run a backup of the data on the iSCSI storage from a backup application running in the guest operating system.
· iSCSI-based storage is supported for backup by the Hyper-V VSS writer when the storage is connected through the parent partition and the storage is used for virtual hard disks.
Whether a backup is performed online or offline depends on whether the backup can be performed without downtime.
An online backup can be performed with no downtime on a running virtual machine when all of the following conditions are met:
· Integration services are installed and the backup integration service has not been disabled.
· All disks being used by the virtual machine are configured within the guest operating system as NTFS-formatted basic disks. Virtual machines that use dynamic disks or the FAT32 file system prevent an online backup from being performed.
· Volume Shadow Copy Service must be enabled on all volumes used by the virtual machine with a specific configuration. Each volume must itself as the storage location for its shadow copies and that mapping must be available to the Hyper-V VSS writer. In other words, the shadow copy storage of C: should be on C:, the shadow copy storage of D: should be on D:, so on an so forth.
If an online backup cannot be performed, then an offline backup is taken. This type of backup results in some degree of downtime. A variety of factors can affect the time required to take an offline backup. If the virtual machine is running or paused, it is put into a saved state as part of the offline backup process. After the backup is completed, the virtual machine is returned to its existing state.
With a "tip of the cap" to creators of Planet V12n blog, we've just launched a blog and Twitter aggregator for virtualization. It's called VirtualizationFeed. Like Planet V12n, this site aggregates virtualization blogs from lots of sources. In this case, 18 independent bloggers and 14 Microsoft bloggers on either TechNet or MSDN. I'm sure you'll recognize many of the names/blogs ... and maybe you'll discover new ones. For example:
The blog aggregator will show a short (200-character) excerpt from the blog and point you to the blogger's site for the full post.
In addition to blogs, you'll see a tab for Tweets. VirtualizationFeed.com is also a Twitter aggregator (supplements the use of Twitter's useful search tool) that pulls all virtualization-related tweets.
I'm told that someone is working on a Vista gadget so that you can view the blog aggregator's RSS feed from your desktop. I'll post the download link once the gadget is available. Otherwise it's self explanatory. Check it out.
Patrick
Hello, I’m Kevin Knox, VP Worldwide Commercial Business at AMD. Microsoft invited me to do this guest blog post in conjunction with our sponsorship of Microsoft’s “Get Virtual Now” event on Sept. 8.
I always find it interesting to hear people singing the praises of x86 virtualization and talking about how this recently introduced technology is already revolutionizing the industry. Fact of the matter is that virtualization technology was originally introduced for the purpose of time sharing on mainframes in the early 1970’s. One could probably trace the roots of x86 virtualization to the early 90’s when IT managers finally realized that people’s desks and wiring closets were probably no place for servers and started to relocate them into the datacenter. The next step in the evolution of x86 virtualization was a few years later when IT managers realized they could safely run multiple applications on a single server simultaneously. And if you are reading this blog, you can probably figure out what came next…..running multiple versions of an OS on a single piece of hardware, or what has affectionately become known as virtualization. While certainly an interesting history, there are two major happenings on the near horizon that I believe will permanently change the face of virtualization.
First is the notion of virtualization becoming “intelligent.” Think about the possibilities when an enterprise infrastructure becomes smart enough to essentially manage itself based on a predefined set of rules, policies and current application requirements. Application A runs for 1 hour and when it is finished, it shuts down, creates a new virtual session with ½ its resources to start a nightly backup and it reallocates the other ½ of the resources to another session getting ready to kick off a major batch run. While just an example, hopefully it shows the possibilities of what some are calling Dynamic Provisioning. In the end what this will enable is the elimination of human capital, which for most IT shops is still the largest budget item they have. No longer will masses of IT people be needed for starting, stopping, configuring and managing virtual sessions. On top of that, by optimizing the software environment in such a way, less, better utilized hardware will result.
The second and probably more important happening is Microsoft’s entry into the virtualization space. While Microsoft has invested in a variety of virtualization areas over the years including partnerships, acquisition and new technologies, they now seem to have a clear strategy and roadmap for delivering broad virtualization technologies across the enterprise. This is not to suggest they will be better or worse than anyone else, but rather having Microsoft in the game will help drive innovation, push the competition and allow for a more competitive marketplace, which ends up being a good thing for end users. Microsoft’s years of experience in operating systems and infrastructure software will also help enable new features and function and enable emerging technologies such as Dynamic Provisioning. One thing I can assure you from my years at AMD is that competition drives innovation and value for the end-users, for the ecosystem and the industry as a whole.
To illustrate this, consider the following advances that AMD has driven into the x86 processor over the past 5 years together with greater choices in virtualization software that are starting to open the doors for IT departments, enabling them to run a larger range of workloads on virtualized servers. 64-bit and multi-core processor technologies now provide the robust platform needed for memory-intensive virtualization. More advanced power management capabilities of the processor are helping to reduce power and cooling costs. AMD-V technology, hardware-enabled virtualization inherent in our processors, is enabling software like Hyper-V to handle the most demanding applications.
So what does the future hold? Well I can’t tell you that, but I can assure you we are in the midst of the most interesting time in the history of x86 virtualization from both a competitive as well an innovation perspective. As Bob Dylan once sang “The Times They Are A-Changin.”
Kevin Knox is Vice President of Worldwide Commercial Business at AMD. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.
Today we announced some changes to server application licensing and support policies related to running MS server apps on top of anyone's hypervisor. Several folks have written or blogged about it, here are some:
Chris Wolf (Burton Group)
Virtualization.info
NetworkWorld
Windows IT Pro
Thoughts on application mobility licensing
As of Sept. 1, 41 Microsoft server applications are covered, both per-processor apps and server/CAL applications that are available via Volume Licensing [Enterprise Agreement and Open Agreements]. In essence the 90-day mobility rule is removed for these applications. With these changes, both licenses and software can move more freely across servers in a server farm [up to two data centers each physically located in a time zone that is within four hours, or within EU and/or EFTA].
Here are some examples of how the licensing might work for you:
· A customer has a server farm with 8 4-processor servers, running a total of 4 copies of Exchange.
o Under the old rules, they would need to either manually move the Exchange instances to another server that is already licensed for Exchange, OR they would need to license all 8 possible servers for Exchange.
o Starting Sept. 1, they will need to have a license for each running instance (4) and those licenses can be moved from one physical server to another as needed.
· A customer has a server farm with 8 4-processor servers, running a total of 4 instances of SQL Server Enterprise Edition under the per-processor model.
o Under the old rules, the customer would need to manually move an instance from one licensed processor to another, or they would need as many as 32 licenses (8x4)
o Starting Sept 1, the customer will need a maximum of 4 licenses. Because Microsoft allows unlimited instances on a processor licensed for SQL Server Enterprise Edition, the customer could have as few as one license if all 4 instances are always moved together.
A few other points:
Thoughts on technical support
As of today, you'll see that we're expanding tech support policy for (initial) 31 server applications for customers that run these apps on WS08 Hyper-V, Microsoft Hyper-V Server or any other validated hypervisor (type 1 or 2). The nut of it is ... customers will be able to get the same level of tech support for virtualized workloads that they get today with non-virtualized workloads.
The kicker here, and where many journos reported inaccurate information, is that 3rd-party vendors' hypervisors must first pass the validation test before customers can get cooperative support from Microsoft and that vendor. For example, it was reported that VMware signed an agreement to participate in the Server Virtualization Validation Program. That much is true. However, it doesn't mean that cooperative support is now in place. First, ESX Server must go through and pass the validation test. Once validated, they'll be added to KB article 944987, where we list "support partners for non-Microsoft hardware virtualization software." Today only Novell is listed, and that's due to the broader technical collaboration agreement in place between the companies.
The other thing to note is that the server application teams have posted configurations that will be supported running on validated hypervisors. For example, the Exchange team posted a blog about their policy, which can be summarized as:
The Sharepoint team blogged about their policy today and posted an FAQ here.
Let us know if you have questions. Cheers,
Patrick O'Rourke
Hi, my name is Dominic Foster, Chief Technology Officer at MaximumASP, which is a web hosting and IT services provider that sells/markets/services Microsoft technologies to the SMB business market.
A core element of the MaximumASP business model is a complete commitment to exclusively offering hosting solutions that deliver Microsoft technologies. This overall market and product focus has been the leading success factor in our growth over the past eight years.
Server consolidation was a very appealing aspect of virtualization - it quite simply allows us to reduce the amount of physical machines in our datacenter. This consolidation gives us greater utilization on our hardware and reduces costs on machines, licensing, and power. Reducing the amount of our physical servers in the datacenter also means less power and cooling costs and lower overall total energy consumption.
Virtualization also gives us the flexibility and scalability we need to make smooth changes to the underlying infrastructure with minimal impact on systems or applications.
During our research we found the TCO of Windows Server 2008 Hyper-V with System Center to be a less expensive solution than a similar toolset from VMware. Being completely manageable by System Center allowed us the flexibility to use the same management tools for both Hyper-V and physical servers. An added benefit was the consistent Microsoft look and feel for System Center so there was not the typical steep learning curve associated with a new toolset. Strong support from Microsoft and the MVP program facilitated our quick adoption of Hyper-V.
It was important for us to be involved in the Hyper-V Go Live program to get access to resources and contacts that could help us bring our solution to market faster. It was also an opportunity to connect with Microsoft to allow us to get our questions answered, and brainstorm our ideas. We've been able to get our questions answered, and/or routed to the product team. We've been able to connect with technical, and marketing Microsoft contacts that help us further develop relationships in one of our 'core' competencies and focus areas.
In May of 2008 we dispatched a virtualization survey to 2,313 customers and received a 10% response rate. The response was overwhelmingly positive with 79% of respondents saying they would consider running their web applications on a virtual server.
Here are some open ended comments from the survey:
· "Being able to store a configured image of a machine and just click-and-drag that image into reality is earth shattering. Deploying virtual SANs, load balancers, and firewalls for pennies on the dollar would give my company a real advantage in our niche."
· "The line between software and hardware is becoming increasingly blurry. Companies like mine must leverage these innovations or yield market to those who can."
Virtualization will continue to be a major focus for us, not only for reasons of consolidation and reduction in operating costs, but also to enable more rapid deployment of new application servers, reduced down time for server maintenance, to reduce driver dependencies for new hardware, and the assistance in disaster recovery planning.
Dominic Foster, Chief Technology Officer, MaximumASP
www.maximumasp.com
UPDATE (August 11th)
Thanks for the comment by wanderson. Microsoft has asked me to update my original guest post so I can address Wayne's comment.
There were a few key features that we required when evaluating solutions. While VMware has a rich feature set in VI3 Enterprise, some simply were not needed. For our environment Hyper-V managed by System Center Server Management Suite, not just SCVMM, was a perfect fit. Some of the key features we needed were:
64-bit Support - Both support this.Migration - While quick migration from Microsoft does incur downtime when the machine saves state then shift the LUN owner, we found the period in our environment rather short. It is obviously not a Live Migration like Vmotion, not yet anyway, but we did not require this.
High Availability - Both support HA/Clustering
Online Backup - We wanted to provide two methods of backup so we decided on Volume Shadow Copy Services from the host and System Center Data protection manager inside the guest.
During the evaluation period the Cost was structured as the following:
Vmware per Server
Virtual Machine Solution VI3 Enterprise 2 Processor x $5750
Management Server (VirtualCenter): $6044
Hyper-V and System Center per Server
Pricing for Server Management Suite Enterprise $1290
Vmware pricing from http://www.vmware.com/products/vi/vc/buy.html
Microsoft pricing from http://www.microsoft.com/systemcenter/en/us/management-suites.aspx#Anchor3
I have omitted the cost of the Windows OS since in either scenario the cost is the same.
In addition we currently use pieces of System Center in our environment and would have to also purchase it if using VMware.
In closing I have to say if the cost of the products during our testing were equal we would have chosen VMware. It has been the industry leader, and is more mature in its development and feature set. However from the numbers above you can see there is a drastic difference in the pricing of the two products and our requirements did not warrant the additional cost.