Monday, August 30, 2010


In a recent article in Network World, the CEO of Riverbed said that, "...Cisco can have Layer 2-3." I know the implication of this because I'm an IT Professional and have a background in networking. As I'm reading the article, I'm thinking to myself, "How do you sell something like this (a reference to any layer of the OSI model) to management?"

The OSI model gives us a way to explain the various parts of networking, also referred to as the network stack -

Layer 7 - Application
Layer 6 - Presentation
Layer 5 - Session
Layer 4 - Transport
Layer 3 - Network
Layer 2 - Data link
Layer 1 - Physical

So when Riverbed says that Cisco can have layers 2-3, they mean they dominate the data link (layer 2) and network (layer 3) layers of the networking stack. Riverbed's contention is that layers 4-7 are WHERE the future of connectivity will reside.

All of this rolls into the larger discussion about "Cloud Computing" that is going on in the business world. Where do your apps reside? Where do your servers reside? Where does your infrastructure reside? Cloud computing is a shift away from on premise (a server sitting on your local area network) to Anything-as-a-Service (apps, servers, and even infrastructure sitting on the Internet and a wide area network).

I like the quote, "...importance of having knowledge workers connect to their applications effectively, cheaply, globally, seamlessly, 24-7..." because this is the ideal of what connectivity should be. Is this type of connectivity possible in the realm of net neutrality? What happens if a network carrier downgrades traffic from a Riverbed device because that same carrier has an agreement with Cisco to prioritize traffic from Cisco devices (such as an ASA 5505) above all else? Something to think about because it is not out of the realm of possibility (see recent 'agreement' between Google and Verizon regarding net neutrality).

If you run a business and have an office, then you'll still need a local area network and that means a gateway device (router and/or firewall), network switching gear (LAN switches), and wireless connectivity (wireless access point). I'm not a big fan of the all-in-one devices sold to consumers for their home network setup with the wireless access point built into the router - I like to keep those two functions separate. In the office, a user must be required to identify themselves & authenticate locally before logging onto your network - "trust but verify" (an excellent cyber security principle). So even if all of your apps and file servers are "in the cloud" the business will still need local resources for things like printing and authentication to the local network.

Something else to think about is integration - how do you get two software-as-a-service offerings to integrate if they are from different vendors? How do you integrate your Internet-based CRM with the online version of your Accounting software? I have personal knowledge of this issue and will tell you that it is a major hurdle. Vendors end up pointing fingers at each other and tell you to get rid of the other guy and go with their all-in-one offering even though it doesn't have nearly the features of the other product. How do you integrate email, CRM, and accounting software when all of them are software-as-a-service?

Of course, that's just my opinion. I could be wrong.


No comments: