Cloud/SaaS/SDN

Virtualization Trends

Sharon Barkai, Co-Founder of ConteXtream, a BGV Portfolio company shares his views on what 2014 holds in regards to Virtualization (Article reprint from the Virtual Strategy Magazine). When looking ahead to what 2014 holds in regards to virtualization, it is perhaps wise to anchor the outlook on some of the natural trends we have been witnessing during 2013. Mobile carriers connect our personal devices, (including smartphones, tablets, cars, watches, etc.), to the cloud content we can’t get enough of, (Facebook, Netflix, Amazon, etc.). As the proliferation of these devices continues and our content choices climb, the focus on the virtualization of mobile and content network service provider networks will remain a hot and relevant topic. This is the most competitive, dynamic and demanding networking environment today, and since telcos will require innovative solutions to ensure survivability and success for both now and in the future, we will be sure to see a strong showing of virtualization actions in 2014. One underlying key assumption is that network providers such as Verizon, AT&T and Comcast will eventually have to structure their infrastructure in a manner similar to that of Internet giants like Google, Amazon and eBay. This is eminent because carriers operate their networks between the proverbial “rock” of users’ insatiable mobile apps demand, and the “hard place” of abundant, low-cost mobile Internet supply. If Internet giants can scale cloud capacity based on cheap, commercial off the shelf hardware (COTS), and can quickly roll out functionality based on easy-to-scale, core map-reduce software, then carriers have to be able to do exactly the same for the public network functions they provide. That requires virtualizing their network. But the questions are, how much of this will actually materialize during 2014, where exactly do we stand in terms of ramp-up, and what is the wave length of this trend? We went into 2013 as an industry with multiple carrier network virtualization trials and (Proof of Concept) POCs already under our belt, and at least one, large-scale U.S. Tier 1 deployment of mobile functions virtualization, aka Gi LAN or CloudGi, with a few more well underway to production. It’s difficult to pinpoint how many other large scale mobile network virtualization production rollouts were completed during 2013, but one thing is for certain; there are only more ahead of us. What’s more, is that we spent, as an industry, a full year on European Telecommunications Standards Institute (ETSI) and Network Function Virtualization (NFV) discussions following the release of the initial multi-carrier NFV white paper. We also were able to witness major progress around carrier-scale network virtualization that was achieved in the various Internet Engineering Task Force (IETF) network virtualization overlay working groups. However, industry wide architectural discussions and initial network virtualization deployments are one thing, and a comprehensive transformation of carrier cost structure and network operations to software models, is quite another. The technological gist of software defined in-network, map-reduce needed to enable carriers to make the shift to global software cloud and cost models, are more or less clear. These carrier Software-Defined Networking (SDN) technologies work, despite the fact that unlike Internet providers, carriers do not typically write their own apps and are extremely geo-distributed in nature. The question is when will these actually make their destined profound impact and transform telecom to standard IT. The software point innovations that can drive this structural shift in telecom forward are already out there and range from advanced optimizations of chatty-connections and the content itself, to smart monetization, custom enterprise tenant applications, and joint marketing that enables revenue-sharing  with over the top apps. So how much of this vision can we expect to come to fruition during 2014? Quite a bit I suspect, or rather,  hope. I expect at least one major such production project among each of the Tier 1 carriers. This can be attributed to the competitive pressures between carriers and over-the-top (OTT) providers. If there is an optimization or monetization innovation that a carrier can “weave” into the network more quickly or efficiently than their competitors, providers may find themselves knocked out of the race. One specific mobile virtualization transformation that generates a lot of curiosity and buzz is that of the virtual evolved packet core or vEPC. These specific carrier functions help connect the radio frequency (RF) network to the Internet by making moving objects seem as if they are stationary to the rest of the world. If IP is the heart of a mobile carrier, than the EPC is the brain. In this area, many industry players expect lots of trials and also some limited rollout in 2014. However, this specific transition is a bit of a catch 22 that will take a few iterations to get out of lock step. This has to do with the essence of mobility itself in which any user may be hooked to both virtualized and non-virtualized EPC components. A migration of a given region to vEPC cannot contain today’s on-the-go consumers. It will likely take a few backward computability bootstrap moves over the next couple of years, but we will eventually get there. To sum it up, I predict that when we look back at this period of evolution among carrier networks, we will be quite amazed at the monumental shift that took place and the speed at which it occurred. We will be equally impressed at the evaporation of the last resort of turn-key “mainframes” triggered by network virtualization as we were at the IP, broadband, and smartphone major revolutions. Right now we all remain students of virtualization, with big ideas about the future and its potential. While we’re no longer “freshman” and our majors as specific players are decided, there is still much left to discover. But “graduation day” is not 2014 but may be closer than we think. For more information read ConteXtream Predicts SDN & NFV Market Tipping Point for Telcos in 2014 @ http://www.contextream.com/news/pr20140122.html