Excerpts from the Interview

On Storage Enabled Network
What we're after is taking SAN and move it from the Data Center onto the Wide Area Network.We want to pull it out of the data center and not make it the data center owned thing. To do that we are leveraging a large toco networking structure. 

We provide a lot of customers wth LANs today. What we basically want to do is make big storage farms that a whole bunch of users can come and access and we can run those at scale because they're big and centralized. You put a bunch of users on it, then lower the overall price of storage from each of those individual users. We are in process of doing that in the South East US where we are deploying central storage resources, we are putting software in place etc. We've reached a point now where we need to work some more with the vendor community and we need more development on the software side to further automate a lot of the process, we need more development on the hardware side, to enhance shareability of resources across multiple user and multiple applications. We believe this is where SAN is going and that its going to move on under the WAN. 

You are going to have both data been created and pushed out under the WAN and users from the other side accessing that data off the wan. That is in essence, what we mean by Storage Enabled Network.

On the standardization efforts.

I think theres been a lot of good work to date on things like fibre channel for instance. We are starting to get to some basic standards, but we still got issues around. For instance ,in the case of data replication, if a customer has a EMC disc array on his side, I have to buy an EMC disc array on my side and then I have to buy EMC software too. Ideally I shouldnt really care about about what the customer has. I can have whatever is the best to have on my end so that I can support shareability across multiple customers. 

On interoperability issues.

I may make the choice that because the customer needs this much performance and reliability he'll need a Hitachi, but I may also say that I dont need great performance on a particular data so I dont need to spend exorbitantly on this. So I may settle for some other option which I can get for a fraction of the cost, and that to me, is where we really want to get.

On the evolution of SAN.

Well nothing ever develops as fast as you want it to. Especially when you are pushing the envelope to the edge, you would want to see change happen as quickly as you can. My outlook is that we are probably 3 - 5 years away from what we are trying to do and that's ok. The important thing now is that we need to get to an agreement head directionally in one way and start taking steps towards that. If it takes 3-5 years that's fine.
I think there are two forces at work here, one is the underlying technology that supports it and the other is, in going out and working with the customers and get them to understand, whets possible. And customers don't change over night. Most of them already have big investments and they'd look to change the way they are doing things as they begin to replace, upgrade and change out the investment they have already made. So, even if you have all the technology available next year, we couldn't start to capture the markets. So, we have work to do, not just on the side of safe standards and develop underlying technology to enable it. We have a lot of work to do with education in the end user community.

On educating the end-user community and the strategic areas that need focus.

I outlined a framework in my presentation yesterday that we put a US patent on and within that framework, I think right now there's a fair amount of work that I know is happening in the networking side and there is also a fair amount of work underway on the hardware platform side too. 
The operating software layer needs a lot more investment. Security is going to become a bigger issue, it hasn't become a big issue till date because its still in the data center. But, if you start integrating all the data centers, security is going to become a bigger issue and a lot of work is going to be required in that area. There hasn't been really much at all done on quality of service. 
Software required to ensure effective service delivery and quality, because not all data should be treated the same and we have to get to the point where we can determine what data is important, and what data is less important, what data is new, what is old, and how to manage that in a 'quality of service' framework. Not much has been done there. And then probably, our biggest area is around putting intelligent devices in the network to route and move data around into the right places. 

On his role in BellSouth 

I work in BellSouth's new business ventures area and so, we look at a number of new businesses to get launched. Storage was a very important one for us. I came out of there and into storage, to get it launched, framed up and in place and so now I look towards other ventures like grid computing, Voice Over IP, security and a number of thing we're doing to build applications that will run over a network. 

Mr. Zimba holds a Bachelor’s Degree in Economics and Policy Studies and a Master’s Degree in Public Finance from the Maxwell School at Syracuse University. His other interests include golfing, sailing, scuba diving, and real estate investing. He lives with his wife and two children in Atlanta, Georgia.