At long last, the wait is over! For the past month or so Cisco has had all of us industry watchers holding our breath for a big announcement that Cisco claims will revolutionize the Internet. The website that has been promoting the announcement has had a "Y2Kesque" timer counting down the days, hours an minutes to the launch. Over the past few weeks there has been a wide range of guesses as to what this announcement is. I've heard everything from a new set top box with Apple TV like capabilities to Cisco building out its own fiber network to the launch of Starent at Cisco to enable 4G connectivity. My guess was actually Starent and 4G but I too was wrong. So what was the announcement?This morning Cisco announced its CRS-3 (carrier routing system) which Cisco touts as the foundation for the next generation Internet. Now for those who remember, around five years ago Cisco launched its CRS-1 among similar hype. The most immediate question that comes to mind is "what happened to CRS-2?" When I asked Cisco about that they claimed they "innovated right past CRS-2" which is one of the most marketingish claims I've heard in a while!
All joking aside though, after looking at the specs on CRS-3, I have to admit, it's a beast. 322 Tbps of capacity (approximately 3x CRS-1, hence the name CRS-3) and it's a 100 Gig-E ready product. This is true 100 Gig-E too, not a multiplexed interface like many of the 40 Gig-E offerings on the market today. The product also has a tight coupling to both UCS and Nexus to enable service providers to offer more competitive, cost effective cloud services. Additionally CRS-3 is IPv6 ready to prep for the explosion of mobile handsets. Lastly, Cisco claims that CRS-1s can be field upgraded through replacing the line cards that are in it. While the product is somewhat interesting (sorry Cisco it's hard to get all that excited about a box) it's really what's driving this level of bandwidth that's the really interesting part.
What Cisco claims is driving the bandwidth is similar to the drivers that I've been talking about for the past couple of years. Cisco views the convergence of mobility, cloud computing and video as the key driver to bandwidth growth. In fact they claim that sometime in the future (they didn't give a time frame) we would all be using about 15 TB of bandwidth per month. The vision that I've been using to frame my research is built on the convergence of cloud, mobility and social media where social media is broader than Cisco's video-only vision. While one can argue the number of 15 TB and whether that's accurate or not, the fact remains that whatever bandwidth is out there, we'll find a way to use up.
While I see the three trends that Cisco has outlined as continuous long term drivers, I do think the short term driver from the telcos will be hosted UC services. This includes voice, videoconferencing, Telepresence, videosharing, unified messaging and all the other great bandwidth intensive applications that we cover as an industry.
I've talked to many network operators over the past few years about what their plans were with regards to cloud computing services as well as what their strategy was around mobility, and in both cases, UC was the unifying factor, which is logical. The heritage of the carriers is communications so the best way to leverage these new business models is to deliver communications. A carrier's ability to compete for hosted UC services is much greater than its ability to compete in, say, cloud based virtual servers. I also think the carriers hold the keys to the explosion of videoconferencing and Telepresence since they're really the only ones that can deliver the "network effect" to video by connecting all of the existing endpoints to one another. Without the carriers' involvement, video will continue to flounder as a niche technology where the value is limited because the network of people to communicate with isn't nearly as big as it is for other collaborative applications.
So while I think the announcement of CRS-3 doesn't match the crazy amount of hype that has been created for this by Cisco and aided by the media, it is a solid announcement that provides a strong foundation for network operators to build emerging services. For those that think the box is just too big and we'll never use that kind of bandwidth, remember we've said that before and we've always been wrong. I'm of the opinion now that we can't have enough bandwidth.
