Three client virtualization approaches to make your investment more sustainable

Extended desktop replacement cycle for client computing is not a boon? In the latest Wall Street Journal of Business Technology, Ben Worthen points out that companies will forgo the traditional three-year replacement cycle for desktops, including traditional desktops and laptops.

According to a recent survey of corporate information executives, 46% of respondents postponed 1-2 years to update their company's machines.

Any machine purchased in the last three years should have at least 2G of RAM, which is sufficient for most people's workloads, Worthen said. In terms of data, most 3-4-year old machines should have 40GB of storage, possibly more.

The current workload may not cause problems, but not for the future. For a special example, companies are now turning to Vista or Windows 7 operating systems. Both versions of the operating system require a lot of hardware resources than the original XP system. As a result, there will be conflicts between future operating systems and current hardware resources-and this is a huge opportunity for client virtualization.

Currently, there are three ways to enable client virtualization to help companies that are currently in an economically constrained environment out of trouble:

Using virtualization becomes possible if the current hardware does have some overload due to heavy applications. The technology can put the application back to the server side, providing only the user interface to the client machine. The current machine is no longer responsible for running and storing data for the entire application, but only a powerful client.

If the client hardware is not sufficient to run Vista or Windows 7, then migrating to a Virtual Desktop Architecture (VDI) environment, using a virtualized server to manage multiple desktops is a good choice. There is no need to assemble 4GB of memory and 200GB storage for the terminal machine, and there is no need to extend the resource level on the server.

For example, if you have 10 desktops on a single server, you don't need 40GB of RAM. Because the end-user machine in the terminal is very small in resource usage, frankly, in the case of 99% utilization, only a small portion of the server's resources are needed. In other words, resources may be reused. This is economically very smart to consider, there are mainly two reasons: (1) as a result of time-sharing reuse, you do not need to buy as much as the use of multiple single point of the resource space, each single point needs to have sufficient capacity to support a Vista environment, (2) for the server, in a sense, Buying large chunks of resources is cheaper than buying the same number of personal terminals – the essence of the problem is actually the price difference between wholesale and retail. This tradeoff between wholesale and retail is also limited, and in general, when you buy a large amount of memory to the server, the price will gradually rise.

For those who cannot share a single server, VDI has other benefits. You can put the client machine in the data center, but specify a blade server for it. For a specific user, make the investment in hardware. But the system design of this blade server will still reduce the overall investment. When you extend the storage quantity linearly for each specific blade, you can still save on the cost of power supply, network connectivity, and cooling.

Of course, all three options need to have some kind of interface device in the terminal location-after all, there needs to be a screen to view the results and a keyboard to enter the information. For the first option, the current hardware can be used. For the next two options, an existing desktop can be used. However, a thin client is still a possibility. This also means that the current user can have the opportunity to implement VDI, use existing desktop machines, and a thin client device is available to the new user when they join the company (or when an existing user machine needs to be replaced). The cost difference for a full desktop device and a thin client is enormous. I hear the difference is 200 dollars, for thin clients.

Of course, there are other factors that need to be considered. The ability of the network needs to be checked to determine if it can handle network traffic between the data center and the desktop client. Of course, the ability to dissipate heat to each end user must also be provided in the data center.

On the other side, the reliability of client devices can be greatly improved through central management, and online access for employees of midrange users should not be supported.

I don't know if I've seen some good examples of how many real desktop virtualization implementations are. Visually, desktop virtualization seems to require less cost to run. There are, however, some examples of possible economic benefits in the early stages, but at a later stage it makes no sense. Inertia makes those systems that no longer make sense to remain where they are, because that is what they have been doing. Desktop computing can make a lot of sense, because it brings computing power to the users of every terminal, but only a fraction of the resources of a mainframe. Makes sense by having end users run their own computing tasks. It also encourages innovation, as many and many applications become meaningful in the computing environment of the end user (the Web). However, this does not mean that future benefits can be obtained only by using a powerful computer in front of everyone.

Smart Corporate information executives will see the benefits of client virtualization, especially in the current economic environment. Perhaps the desktop can also extend the service life of 2-3 years.