Archived Content

The following content is from an older version of this website, and may not display correctly.

With all of the announcements at VMworld 2013 around VMware's NSX, it sounds like every problem in the data center can "now" be solved.

I wonder why everyone is so hot on this technology. I viewed the product demo and that did not strike me as impressive so far.

For a typical enterprise (not talking cloud data center!), the challenges of provisioning services could be solved today with the existing software-defined networking (SDN) solutions and APIs out there. It looks like NSX just adds another layer of complexity for those customers with a couple of hundred or a few thousand virtual machines, as routing, switching, and firewall features are replicated at the NSX layer. A modern switch fabric that embraces agility and enables automated VM mobility can be easily be built today. The days where the network was the limit are really over. But one needs to turn the new features on, obviously, and a strategic plan is required here, and the network teams are too often slow movers compared to the system team.

But based on VMware's market influence right now, everyone needs to follow the hype for now ... let's see how that evolves.

Very often this lack of agility in IT has its root causes not in the technology but within the organization and workflows. People hope to overcome that with technology. The network and system teams just don't talk to each other and are not aware of all of the challenges, or the possibilities, that might exist out there. So sitting down at single table with all stakeholders and technology providers, including VMware, can bring solutions to daylight that the teams did not see before.

My own experience talking to customers on a regular basis just confirms that. Sometimes it is easier than people might think. Everyone wants to get to the holy grail of data center operation — the Software Defined Data Center (SDDC) — but I think there are several paths to get there. One needs to carefully choose the one that is tailored to the demand. Becoming aware of what those demands are is the first and most important step that needs to be taken. If I am an enterprise, I might not be interested at all in 16 million different virtual networks. Doesn't add real value. If I am going with multiple hypervisor technologies so I am not dependent on one supplier, I also need to make sure that I can execute my strategy against those multiple hypervisor solutions. And so on...

A lot of changes in the data center and in data center networking are happening as we speak. The customer should keep a calm hand on the rudder to navigate through that stormy weather and select the right technology that suits their needs.