TechTarget Windows Media SearchWinIT.com SearchExchange.com SearchSQLServer.com SearchEnterpriseDesktop.com SearchWindowsServer.com SearchDomino.com LabMice.net Hyper-V: Microsoft s Approach to Server Virtualization E-Guide Organizations are looking to server virtualization to consolidate, increase efficiency and save money. Microsoft s approach to server virtualization is significantly different from that taken by VMware. In this informative E-Guide brought to you by SearchWindowsServer.com and Dell, learn specifically how Hyper-V is designed to meet high availability requirements, how this technology is being used at the hardware level, the role of a 64-bit operating system with virtualization, and more. Sponsored By:
Table of Contents E-Guide Hyper-V: Microsoft s Approach to Server Virtualization Table of Contents: Can Microsoft Hyper-V meet high availability requirements? Server virtualization at the hardware level with Hyper-V Virtualization and 64-bit: A match made in Windows heaven Resources from Dell Sponsored by: Page 2 of 9
Can Microsoft Hyper-V meet high availability requirements? Can Microsoft Hyper-V meet high availability requirements? Danielle Ruest and Nelson Ruest, Contributors The buzz in the industry right now is all about virtualization. Virtualization vendors are jockeying for position and each one touts the features the others are not supposed to have. One of these is a feature every hypervisor should have and one that Windows Server 2008 Hyper-V seems to be without in this version: live virtual machine (VM) migration. Live migration in a virtual environment does not mean migration from one state to another, such as migrating a physical machine to a virtual state. It means moving a VM from one host server to another while the VM is running and delivering service to end users without interrupting the service. In order to move virtual machines in this manner, you need to make sure that each host server has access to the files that make it up. When you move a VM through live migration, you don t want to have to move the files that make it up since those can be considerable in size. Instead, you want to move only the in-memory contents of the virtual machine contents that are stored within the host server s memory. Both VMware ESXi and Citrix XenServer have the ability to do this, and both use the same strategy. Generally, host servers are linked together in high-availability clusters or resource pools. The servers tie into the same shared storage container and because of this, they have immediate access to the files that make up the VM during such a move. This is the first rule of host servers: They must be configured to tie into shared storage in order to provide high availability for the virtual machines they host (see Figure 1). Figure 1 Microsoft s Hyper-V does not support live migration. Instead it supports Quick Migration a feature that saves the state of a VM and moves it to another host. Because the state of the virtual machine is saved, there is an interruption in service, even though in some host server configurations this interruption can be as minimal as four seconds. Hyper-V provides this feature through Windows Server 2008 s Failover Clustering service, where host server nodes are linked together into a failover cluster. These clusters can provide host server redundancy at the site level when two or more nodes Windows Server 2008 can create clusters of up to 16 nodes are linked to shared storage, or at the multi-site level when two or more nodes are joined through WAN links to provide redundant services should damage occur at the site level (see Figure 2 on next page). Multi-site failover is based on each cluster node having the same contents as the other. This is achieved through content replication, a feature that Windows Server 2008 does not offer you get it through a third-party replication Sponsored by: Page 3 of 9
Can Microsoft Hyper-V meet high availability requirements? tool. Despite this, multi-site clustering is available by default for Hyper-V. Other hypervisors do not offer this out of the box. Figure 2 Windows Failover Clustering and Network Load Balancing (NLB) are two features that have been part of Windows Server since Windows NT (though NLB was called Windows Load Balancing at the time). These two features are mutually exclusive, but both provide high availability to some degree. Failover clustering is mostly used for the protection of stateful workloads or workloads that must commit changes to data, such as SQL Server or the Mailbox Server role in Exchange Server. Network load balancing clusters are created for workloads that are stateless or readonly. A good candidate for NLB is the Web server, especially when it is used as a front end to an n-tier application. If you are running Windows workloads in virtual machines and you want to make sure those workloads are always highly available no matter which hypervisor you use, you can and should configure them to use either Windows Failover Clustering or Network Load Balancing. In addition, you can configure non-affinity policies to make sure that each node of a cluster does not reside on the same host server. Then, if a failure occurs either at the VM or the host level, your workloads are automatically failed over without any service interruption to end users (see Figure 3). Figure 3 Sponsored by: Page 4 of 9
Server virtualization at the hardware level with Hyper-V So, is it essential for Microsoft Hyper-V to have live migration? The answer is no, not at this time. Most organizations running Hyper-V as a hypervisor will also run Windows workloads in their virtual machines. By relying on Windows Server 2008 s own internal features, it s easy for administrators to make sure there are no service interruptions to end users, no matter what happens to the host server. It doesn t work for every Windows workload, but it does for most of them, and as a proven technology, it works really well. Server virtualization at the hardware level with Hyper-V Brien M. Posey, Contributor For as long as Windows has existed, applications have been prohibited from communicating directly with hardware. This is because one of the major principles behind the Windows operating system is that it acts as a level of abstraction between hardware apps. Applications never communicate with the hardware directly. Instead, they communicate with Windows, which in turn uses various device drivers to enable it to communicate with the physical hardware. Recently, however, this philosophy has started to change at least when it comes to server virtualization. Let s start with a little history. A look back at Virtual Server 2007 Prior to the release of Windows Server 2008, Microsoft s primary virtualization solution was Virtual Server 2007. Virtual Server used the standard philosophy, that is, that apps were not allowed to communicate directly with the system hardware, taking something of a monolithic approach to server virtualization. Windows treated Virtual Server 2007 pretty much the same as any other Windows application in that the host operating system ultimately retained control of all of the system s resources. That meant guest operating systems all shared system resources, such as memory, network communications, video processing and so on. This sharing of resources is both inefficient and risky. It s inefficient because guest operating systems do not have a dedicated pool of system resources. Instead, the host operating system acts sort of like a dictator, telling the guest OS if or when it can have access to certain resources. Both Windows and Virtual Server 2007 act as a bottleneck for guest operating systems. This is a risky approach because of the way that resources are shared between guest and host operating systems. Suppose for a moment that the host OS had a buggy NIC driver, and that bug eventually made it so the host OS could not communicate on the network. Because the guest operating systems are completely dependent on the host, they would not be able to communicate across the network either. Sponsored by: Page 5 of 9
Virtualization and 64 bit: A match made in Windows heaven Enter the hypervisor With the release of Hyper-V, Microsoft took a completely different approach to server virtualization in that virtual machines are now allowed to communicate directly with the hardware (well, sort of). The exception is disk I/O, which is still coordinated through the host operating system. Guest servers running on Hyper-V completely bypass the host OS and communicate directly with the server s hardware. The reason why Microsoft is able to take such a radically different approach to server virtualization is that Hyper-V is based on some relatively recent changes to server hardware. The latest server hardware supports something called hardware-assisted virtualization. For example, Intel servers offer Intel VT (Virtualization Technology), while AMD has AMD-V. Hyper-V absolutely requires that your server be equipped with one of these two technologies. It is also worth noting that I recently deployed a server that was equipped with Intel VT, but I had to enable virtualization at the BIOS level before I was allowed to install Hyper-V. So what else makes Hyper-V different from previous virtualization technologies? Unlike Virtual Server 2007, Hyper- V is a very small application. This size reduction is due to the fact that it s really the hardware that is doing most of the virtualization work. Hyper-V creates special partitions for guest operating systems. These are different from disk partitions, because the partitions include memory and other system resources. Each virtual machine functions solely within its own partition, which greatly reduces the chances that a failure with the host OS or with a different guest could impact a guest operating system. It also makes virtualization much more secure since the virtual machines are physically isolated from each other. Hardware-assisted virtualization is a technology that is really worth paying attention to. I have been using Hyper-V on a few different servers in my lab, and guest operating systems seem to perform much better than they do in a Virtual Server 2007 environment. In fact, they tend to perform so well that I sometimes forget that they are virtual servers instead of physical ones. Virtualization and 64-bit: A match made in Windows heaven Christa Anderson, Contributor Virtualization is the hot topic of the day. Be it application virtualization, OS virtualization or presentation virtualization; if you can virtualize it, someone s probably slapped that label on it. The thing is, all of these technologies have been around for some time, even years in some cases. Multi-user Windows has existed in various forms since 1992 and became a core part of the Windows operating system with Windows 2000. VMware Inc. has been evangelizing virtualized servers and clients since the company s inception, Sponsored by: Page 6 of 9
Virtualization and 64 bit: A match made in Windows heaven and SoftGrid was talking up application isolation and streaming long before Microsoft purchased the company in 2006. People were buying it, too. So why is virtualization a hot topic now, instead of two years ago? There are several possible reasons. The virtualization features have improved with every release and therefore have become more like working on a non-virtualized computer. Increased interest in environmentally friendly computing solutions has fostered interest in remote access and server consolidation. Still, perhaps the most important reason why virtualization has become such a hot topic is that the infrastructure now exists to support it and make it scale while ensuring a rich experience. Reliable high-speed LANs and WANs are part of that infrastructure, as is 64-bit Windows. In fact, 64-bit Windows is a key part of virtualization because of the one major virtualization bottleneck memory. Let s take a look at the relationship between physical memory (the DMMs you install in your computer) and virtual memory (the place where the operating system stores data and applications in use). In a 32-bit system, Windows can address up to 4 GB of virtual memory. Two gigabytes of virtual memory are shared among kernel-mode processes that support core functions of the operating system, and 2 GB are allocated individually to each user-mode process and isolated from all other user-mode processes. The number of virtual memory addresses available to user-mode processes may appear enormous because each process sees the entire 2 GB area for its exclusive use. But in order for virtual memory to be useful, the memory manager must be able to map the virtual address to a physical location so that when the data is needed, the operating system knows how to go get it. Windows does this through a system of pages that store data, page tables that index the pages and a record of page table entries. Combined, these all document how a virtual memory address maps to a physical location. The 32-bit operating system s method of mapping virtual addresses to physical ones works for up to 4 GB of physical memory under normal circumstances, since the addresses are 32 bits long. The rest of the virtual memory addresses must be backed by an area of hard disk called the page file, which provides alternate storage but is slower than RAM. The issue here is that on a virtualized system, there s going to be a lot of user-mode processes. A single computer may support half a dozen or so users for virtualizing desktops using technology like Microsoft s Hyper-V or VMware s ESX Server. And it may support dozens or hundreds of users for virtualizing applications using Terminal Services. Every user will have his or her own set of applications, and all those applications were originally designed to run on a single-use computer. Virtualization platforms are designed to be as parsimonious as possible with memory, but at the end of the day they re bound by the demands of the applications. Virtualized PCs have an even greater problem than terminal server sessions. The entire operating system must be virtualized to support each connection. Another issue is that virtualization becomes the victim of its own success. Sponsored by: Page 7 of 9
Virtualization and 64 bit: A match made in Windows heaven If the virtualized experience is limited, then people won t like it. But if it s got most of the same features of a nonvirtualized platform, then supporting that takes resources. For example, the new support for monitor spanning in Windows Server 2008 Terminal Services requires more memory than a single monitor because the viewing space is larger. Therefore, you need an efficient virtualization platform with enough memory to back it properly. Although terminal servers have used 32-bit operating systems for smaller deployments, 64-bit platforms combined with adequate processor support and a disk topology designed to reduce I/O bottlenecks will be necessary to support larger deployments. And that s just as true for virtualized operating systems attempting desktop replacement. For this reason, Microsoft s Hyper-V is available only on 64-bit operating systems, although you can install 32-bit operating systems as guests on Hyper-V. There are some catches to 64-bit operating systems too, of course. For one, 64-bit processes use more memory than their 32-bit counterparts, so you ll need to run enough processes to require more than 4 GB of memory before it s worth it. In addition, 64-bit operating systems need 64-bit drivers, which can be harder to find. Still, although they require more planning to implement, 64-bit operating systems are the future, especially since they are more or less required to support the virtualization that people are looking for. Sponsored by: Page 8 of 9
Resources from Dell Resources from Dell Virtualization 101 Using virtualization for testing and development environments What is Microsoft Hyper-V? About Dell Dell Inc. (NASDAQ: DELL) listens to customers and delivers innovative technology and services they trust and value. Uniquely enabled by its direct business model, Dell is a leading global systems and services company and No. 34 on the Fortune 500. For more information, visit www.dell.com, or to communicate directly with Dell via a variety of online channels, go to www.dell.com/conversations. To get Dell news direct, visit www.dell.com/rss. Sponsored by: Page 9 of 9