Quoting Line56
A favorite vignette during the December holiday season is a frenetic wife shopping for gifts as her hapless husband stands idly by. One by one, the wrapped packages pile up in his arms until only his hat is visible behind them. An icy sidewalk often comes into play, bringing all
the packages crashing down around his feet, much to the disgust of the Missus.
Server virtualization can be a lot like that. Virtual machines and encapsulated applications collect on the physical server to make excellent use of its previously underutilized capabilities. It's marvelous technology and smart IT management, so long as you watch for the ice.
A No-Brainer?
Virtualization has become a rock star. New products and expanded capabilities are rushing to market. Industry analysts and news media sing its praises at every turn. For the most part, virtualization technology delivers as promised. An aspect many overlook, however, is the role - or the risk - of virtualization in a business continuity strategy, especially with regards to availability of mission-critical applications. Let's take a look at the whole picture.
From a server perspective, virtualization breaks the bond between the operating system (OS) and the underlying hardware. This means that more than one OS can run simultaneously on a single machine - each instance working within its own virtual hardware. One physical server will support either multiple versions of the same OS, or different OSs running at the same time ... Windows and Linux, for example. With many virtual machines running, each server works more efficiently and to greater capacity. The corollary is that an organization requires fewer physical servers to accomplish its goals, saving on space, staff demands, electricity, and budget dollars.
The potential of virtualization goes beyond just consolidating servers or avoiding new hardware purchases. Since a virtual machine is independent of the hardware it runs on, it can be moved around a company's network to make best use of existing resources, and copies can be saved off-site to protect against the effects of a server failure. This hardware independence means that applications can be recovered more quickly on a virtual machine than on an ordinary physical server; shift the virtual machine over to a different server running the base virtualization platform, and it can be booted up straight away. Further,
patches and security fixes can be pre-tested on a virtual machine that's identical to the instance that's actually running the application.
Application Service Levels
Virtualization has been around for years on mainframes. But in open systems environments it still has maturing to do, especially where business continuity is a concern, which is most everywhere. In the preceding paragraph, for example, we noted that applications can be
recovered more quickly and that patches can be pre-tested in virtual environments. That does not mean, necessarily, that applications will be available without interruption or that transaction data may be lost when recovery or patching actually occurs.
The hubris that can accompany the rush to virtualization can also mask considerable risk. With fewer servers running more applications, the business risk associated with a failed physical server - i.e. the underlying hardware supporting the many virtual machines - increases exponentially. As with our upended holiday husband, all the applications are more likely to come crashing down when you compromise on platform availability and reliability. A Stratus customer expressed it another way during his own server-consolidation project. If I'm putting all my eggs in one basket, he said, then I want a very...strong...basket.
Some applications are not well suited to virtualization. These tend to be programs that require large amounts of I/O or frequent memory access, rather than simply just CPU access. A large database, for example, can require more power and resources than can be delivered within a virtual machine environment. In this case, performance deterioration would far
outweigh the benefits. A better approach would be to run the application on its own physical server.
As a virtualized environment becomes more complex and distributed, greater flexibility through technologies like single sign-on, storage, and security become more critical. They allow users to access their information, no matter where there are or where the workloads move -- all while keeping your network safe and your administration overhead low. Where the workloads go, the security goes with them.
In summary, every savvy IT manager knows that no single solution does everything, notwithstanding vendor claims to the contrary. It comes down to setting clear goals, followed by careful planning, testing, measurement and re-testing. In a virtualization strategy, guarding against downtime - not only for the hardware but for the supporting infrastructure, as well - is key to ensuring uninterrupted application availability and business continuity. Be sure to watch your step.
Read or comment on the original, here.