Tweet

Network virtualization is a method of combining the available resources in a network by splitting up the available bandwidth into channels, each of which is independent from the others, and each of which can be assigned (or reassigned) to a particular server or device in real time.

Each channel is independently secured. Every subscriber has shared access to all the resources on the network from a single computer.

Network management can be a tedious and time-consuming business for a human administrator.

Network virtualization is intended to improve productivity, efficiency, and job satisfaction of the administrator by performing many of these tasks automatically, thereby disguising the true complexity of the network.

Files, images, programs, and folders can be centrally managed from a single physical site. Storage media such as hard drives and tape drives can be easily added or reassigned. Storage space can be shared or reallocated among the servers.

Network virtualization is intended to optimize network speed, reliability, flexibility, scalability, and security.

Network virtualization is said to be especially effective in networks that experience sudden, large, and unforeseen surges in usage.

Most of you in the network arena probably are familiar with the terms virtual LAN or virtual private network. It's a toss-up as to which is actually the older of the two most commonly used virtual networking technologies, but it really comes down to whether you've been associated with remote access or traditional wired technologies.

This past fall, we were kicking around an internal discussion of what might be happening around the industry in network virtualization. At that time, we really couldn't put our fingers on what vendors were doing in this particular area, and there wasn't a great deal of press associated with virtualization in networks as a whole. Sure, there was the traditional news about a new VLAN capability or a new flavor of VPN, but nothing really new and exciting to speak of.

Since that time, we've hearing quite a bit of new developments. There are several interesting technologies that have emerged over the past six months to approach network virtualization issues from a variety of perspectives. With the latest system virtualization advancements, administrators can simply and easily configure a new virtual machine (VM) environment and drop it on any available platform that meets their needs.

As the functionality of that VM changes or demands on its resources increase, the systems administrator can easily transfer that VM environment from one system to another. From the administrator's perspective, that's great, and the process typically takes very little time these days compared with what it took to configure a new system and bring it up to an operational level.

Now, enter this scenario from the network perspective. What happens when that VM environment is transferred to a different physical system and ends up with a completely different IP address? How is traffic going to get rerouted? Since the systems administrator can schedule a VM transfer while it's operational, what's going to happen to the live, real-time end-user sessions? It all depends on whether the network has been updated with the latest network armament that can rise to the occasion.

Routing is of course the biggest issue with these types of VM transfers, along with caching and load balancing.

The network plumbing now has to be VM-savvy to keep up with where the systems are residing at a moments' notice. Routing tables have to be more flexible than ever, and caching elements have to be kept informed when there is a change so they can correctly redirect end-user session traffic from one location to another; otherwise the sessions will become stale, and the quality of the end-user experience will drop dramatically.

The other intriguing development in network virtualization is loosely referred to as a VM tap. When we typically think of network communications, we're almost always thinking of the packet traffic flowing on our wired, wireless or other type of infrastructure plumbing. But what about the communications among VMs or the communications between a VM and its virtual management layer? Vendors are learning that those communications are as important as the traditional packet-level communications.

So they've devised ways to passively intercept those intra-VM communications and channel them outside the VM environment.

Imagine if those redirected intra-VM communiqués could be forwarded over to a no-frills logging system, similar to a syslog server, to gobble up everything thrown its way. What would you pay for that type of business and operational advantage?

Now, what if that same VM communications tap could redirect those intra-VM communiqués over to your favorite network management system, device-monitoring tool or network alert system and could package those redirected communications in a form those systems could easily take in? What would be the business and operational advantages for that type of extended intrasystem capability?