Telstra, Australia’s national carrier, remains locked in a dispute between its shareholders and the Australian government, which proposes to break Telstra up to support its broadband network initiative. Telstra believes it can play a role in NBN, as the network is called, without structural separation. Discussions on how that would be accomplished were proceeding nicely, according to Telstra, until the legislation for separation was introduced.
We are of the view that separation like this is not likely to succeed at its objectives and in fact is more likely to compromise them. In addition, it would clearly undermine the value of Telstra’s shares, which have been sold off to the public for some time through yet another government program. To take steps to devalue them now is not likely to create confidence in any new government process, including NBN.
Then there’s a very important truth, which is that competition always lowers price and always creates overbuild and thus always lowers net industry ROI. There are situations where that is reasonable; some geographies have ample demand density. Australia doesn’t appear to be one of them, according to our model.
If the government’s goal is to create competition in the interior or service portion of the network by making the access network into a shared utility, it should have clear indicators that the ROI in that service portion is sufficient to create real opportunity. Without that, there could be a lot of future failures in the center, which would undermine the profitability and sustainability of the new edge—whatever it turns out to be.
Cries that Juniper must now match Cisco and Tellabs and buy a mobile-core player are ignoring a critical reality — and it’s not that Juniper’s new superchip can make its universal edge portfolio into an Enhanced Packet Core box (though it can).
The real issue with wireless, guys, is that you need a radio network. Through its close relationship with NSN (including metro, which after all is the “core” of Enhanced Packet Core, through its joint venture), Juniper has direct linkage with the radio network. NSN and Juniper actually make a pretty credible total mobile solution provider, stronger than either Tellabs or Cisco, providing that they can make the joint venture work effectively in that light.
NSN also adds professional services, which virtually all LTE deals are likely to mandate. All of this, to us, means that the real question is whether the Juniper/NSN joint venture will work effectively. While both companies blow kisses at the joint venture in their public comments, we’re still waiting to see substantive cooperation emerge.
The head of NSN’s North America operation said the company would be an LTE leader in North America, and said it rather emphatically, to be sure. The comments came at NSN’s analyst event this week, which was a parade of optimism on NSN’s prospects in the market, not only in North America but elsewhere.
NSN’s greatest strength is its professional services business, and second on the list comes its credibility with operators in multi-vendor integration. The company also has a strong service-layer story, though that story is built on tools used by its consulting and professional services staff to create one-off solutions for operators and not productized (at least for now).
The value of professionals services in the U.S. today is based on the procurement zone trend, already formally in place at AT&T, and in process (we’re told) at both Verizon and Qwest, and being considered even by Sprint. This program creates what’s effectively a mandatory systems integration position in each zone, which in turn means that those with strong capabilities in that area can expect to get a prime vendor position.
Juniper’s radical service-layer software and semiconductor architecture speaks to carriers and enterprises
We can’t apologize for the characterization here; Juniper announced a radical combination of an extensive service-layer software system and a new semiconductor architecture, taking the most profound step the company has taken since it was founded.
The new chip is a family, the first member of which is Trio. It is based on a “Network Instruction Set Processor” model that builds software on the device using instructions customized for network behavior control rather than general-purpose instructions, as NPs do. In this respect, the chip is almost like an ASIC, but unlike an ASIC it’s programmable at the primitive NISP-instruction level, so new features can be added right down to the instruction level. It’s this architecture that accounts for the considerable improvements in performance, scalability, power efficiency, etc. that Juniper has demonstrated (through independent lab tests).
The software (Junos Space) is centered on a complete restructuring of the Juniper Junos operating system, and it extends Junos to cover not only Juniper devices but also independent software development, and even a new device client (Junos Pulse) that will provide security and identity management, VPN control, and connection control.
The software side of the announcement is the most critical because it is based on middleware to create what is, in effect, a platform-as-a-service cloud on which operators can build service features and service management components.The software there can then leverage software running on the routers through the old Juniper PSDP program, and through that could even, in theory, be linked to special “primitives” programmed into the new Trio NISP chip.
From a selfish vendor perspective, the most important thing is that Juniper ties the service layer development downward into its devices (through its router development programs like the old PSDP) and even potentially down into the chips themselves. That creates a value circle for them and it also lets operators build differentiation by linking their service solutions tightly into the network, giving better integration, operations, and performance than OTT players could achieve.
Juniper released applications for Ethernet activation, surveillance and monitoring, and problem management in the network. These are built on the Junos Space architecture. This new model is so radical that frankly it’s hard to believe it came out of Juniper, never known either for software or for making game-changing moves. This is clearly one such move, though, because it will at the minimum catalyze the whole service-layer marketplace.
Juniper also announced an expansion of its IBM OEM deal to include the SRX, which offers security control and falls under the Junos Space software umbrella at least in a development sense.
Finally, BLADE Network Technology announced it would license Junos for blade server switches, possibly the first porting of a network OS to another platform. Just assimilating all of this will no doubt create some headaches, but early indications from operators suggest they’re very interested, and there is also surprising early interest among large enterprises, particularly in the cloud computing potential.
Juniper has announced an OEM deal with Dell that will have the latter re-skinning the MX, EX, and SRX products and offering them as a part of the Dell PowerConnect brand. The deal is a validation of the truth we articulated this time last year: Data center IT is now powering data center networking from an enterprise political perspective.
That means network vendors need an IT hook, and Dell offers Juniper another one (Juniper has a deal with IBM). One interesting slant on this is that Dell has other deals with Cisco and Brocade, and both of those make sense given the strong position those companies have in data center and storage networking. But Juniper is a relative up-and-coming.
Since Juniper made a big deal of its Stratus fabric for the data center at its analyst event in early spring, it may be that Dell sees Juniper as having a strengthening role. Juniper is also a leader in financial applications that require low latency, which is a sector every data center player is interested in. Dell and Juniper will also partner on Data Center Bridging, an extension to Ethernet that provides the lossless transfer that’s needed for data center storage and virtualization applications. DCB is also something that some operators are looking at as a cloud data center service.
Verizon delivered a classic US-telco story in the third quareter: Wireless was doing very well, telco TV was doing better, enterprise services were under price pressure, and wireline losses continued.
For Verizon, the net was better than Street expectations. Most encouraging were revenue gains of more than 10%, though about half of that was from the Alltel acquisition. Customer churn was up slightly in wireless, and DSL broadband subscribers continued to decline, but the latter was more than offset by broadband FiOS growth, suggesting that Verizon FiOS is attracting broadband users beyond its own DSL base.
The company promised investors that LTE would not generate a big capex surge, a relief given that the large up-front FiOS load is finally tapering down. In all, the story was good but not great. You can see that Verizon, like AT&T, is spending a ton of money on wireless and making a big bet it will be able to capitalize on that investment. We think that in the near term, that is almost certain. But the long-term viability of wireless depends on the ability of the operators to create value beyond voice and Internet.
Verizon is apparently planning a total smartphone blitz for the holidays, including the already-discussed Droid from Motorola, but also according to rumor, a new HTC Android smartphone and perhaps a third model as well.
The Verizon moves are, we think, are calculated to make things complicated for AT&T and Apple with the iPhone for the holidays. A multiplicity of models at various sizes and price points makes sense at this stage in the market, but there is no question Apple still has the cachet that other smartphones have so far been unable to match.
We think all of these smartphone wars are a signal that whatever the FCC might do with wireless net neutrality, operators are committed to the mobile web.
The FCC isn’t “considering” net neutrality rules as a formal order; they’re “proposing” specific actions. That’s the meaning of what happened yesterday at the public meeting. The FCC approved its net neutrality Notice of Proposed Rulemaking (NPRM) by unanimous vote, but Republicans dissented in part, just enough to keep alive the partisan bickering that’s characterized Washington and to reflect their concern with some of the points.
If you look at the document closely, you find it really consists of four parts. One codifies the original principles of net neutrality that were published in 2005. The second requires that providers treat lawful traffic in a non-discriminatory way and publish any traffic management policies. Both these sections (which are subdivided further in the NPRM) are largely accepted by all. The third element asserts the FCC’s position that these principles be applied across access types, meaning wireline telco, cable, and wireless. This raises some ire in the industry, and also with some Republicans.
The final point, in our view this is the most controversial, asks for comment on the specialized services that are IP-delivered and share the broadband pipe with the Internet. The FCC wants to know how to define these and what rules (including the six defined in the first and second parts we’ve outlined) should apply to these special services.
This last issue is pivotal because should the FCC decide these services are also to be regulated in some way, they could impose access-sharing requirements on telco IPTV offerings. We think it will be months before any order comes out of this process, but just the fact that the FCC is thinking about the third and fourth issues here could force operators to consider what would happen to their mobile and wireline walled-garden services if access connections had to be wholesaled to competitors at a fair price.
Regulatory issues are never covered well in the media, and the public interest in this one will likely make it worse. Make no mistake; the key to the future is in the single issue of those managed or special services!
An interesting piece on the evolution of the Internet shows pretty convincingly that the classic model is being replaced by a new one where content portals and content delivery networks (CDNs) peer with access ISPs, eliminating more and more of the classical notion of an Internet backbone.
We’ve talked about this trend often. Experiences are a resource-to-user connection and not a user-user connection and thus don’t demand universal peer connectivity. While in theory, there are many resources available on the Internet, the practical reality is that most of them are found at statistically insignificant rates and contribute nothing to value and traffic—today.
The challenge we now face is whether the profit-driven evolution toward content concentration and experience networking is going to change the nature of the Internet. Is it becoming a big TV network, in effect? How much of what broadband policy is trying to preserve in the way of innovation and flexibility is already being lost to a quiet restructuring? We aren’t railing against the changes, only against the fact that we’re not facing them squarely and talking about the consequences.
Google, for example, is building the most connective resource center on the planet, and with that, it has a jump on any new application, not to mention the older ones. All of the talk about how Internet traffic is growing (Cisco) or how operators’ backbones are expanding (AT&T, Verizon) treats the Internet as homogeneous. It’s not, and how its structure is changing is critical to everyone.
Cisco has announced a new generation of its popular ISR platform (G2, fittingly), and also announced an initiative/architecture called Borderless Networking. The ISR upgrades are performance enhancements to the earlier models based on what Cisco calls the “Service-Ready Engine” that can support Linux applications directly, not through the older AXP insert card.
Borderless Networking is harder to pin down, however. It appears to be what a Cisco PR video calls a “recommitment” of Cisco to some core technologies rather than a new announcement. But it is possible that Cisco will offer something new and substantive there. From the positioning, it appears to be a service-layer strategy focused on creating an “IT control plane” from Cisco’s data center and virtualization technology. Whether it’s real or slideware is the question.
Cisco often makes announcements like this to anticipate announcements by competitors, and a number of them may be planning something in the service layer area within the next month. As we’ve noted, this is a critical area, and if Cisco can create a credible “IT control plane” based on data center virtualization, it could have an impact in the space. Recent trends within Cisco management and organization, however, seem to suggest a de-emphasis on software products and on network abstraction and management, key ingredients in a service-layer strategy.