Does SDN Need a Faster Version of TCP/IP?

The transport layer of the protocol stack has direct implications for use of software defined networking, many would argue, and therefore creates a need for SDN-optimized versions of TCP/IP.

So it is that the new OpenFastPath  Foundation, founded by Nokia Networks, ARM and industrial IT services player Enea, seeks to create an open source TCP/IP stack which can accelerate the move towards SDN in carrier and enterprise networks.

AMD, Cavium, Freescale, Hewlett Packard Enterprise and Linaro also are supporting the effort to create a standardized, accelerated TCP/IP stack that provides top performance for SDN-ready network functions.

For network operators, the new fast-path stack should lead to reduced IP latency, higher capacity and thus faster packet forwarding, and lower implementation costs.

Public Cloud Services Market Will Grow 19% in Middle East, North Africa in 2016

The public cloud services market in the Middle East and North Africa (MENA) region is projected to grow to 19.3 percent in 2016 to total $ 880 million, up from an estimated $737 million in 2015, according to Gartner analysts.

Business Process as a Service (BPaaS), the largest segment of the cloud services market in MENA, is expected to grow 7.3 percent year over year  in 2016 to reach $266.2 million.

The fastest-growing market, cloud management and security services, is projected to grow 29.3 percent in 2016, year over year.

Gartner predicts that in 2019, total public cloud services spending in the MENA region will rise to $1.45 billion, with Software as a Service accounting for 22.6 percent of the market.

followed by Cloud Management and Security, growing 21 percent, and Infrastructure as a Service growing 19 percent in 2019, year over year.

Global PBX Market Up in 3Q, Down for Year

Global sales of pure IP, hybrid and TDM PBX enterprise telephone systems totaled $1.6 billion in the third quarter of 2015, up 3.9 percent from the previous quarter, while unified communications solutions grew 5.4 percent, according to Market research firm IHS.

On a year-over-year basis, the PBX market was down seven percent in the third quarter, while the unified communications market was flat.

IHS predicts the worldwide PBX market will grow at a compound annual growth rate of -0.1 percent between 2014 and 2019, reaching $6.9 billion in 2019.

Will Linear Video Revenues Drop 66%?

Skinny bundles costing $30 are the future of linear video linear video, says CBS head.

If an average linear video monthly bill is $90 a month, and the future is $30 a month, average revenue per account will drop 66 percent.

That illustrates, after similar devlines in voice revenue, why all service providers are searching aggressively for new revenue sources.

If an average linear video monthly bill is $90 a month, and the future is $30 a month, average revenue per account will drop 66 percent.

At AT&T, for example, voice revenue is only the third-largest revenue source, representing perhaps 22 percent of total revenue.

source: CableOne

source: Spirent

source: CableOne

Xfinity Wi-Fi is Ready for Wholesale. Are Buyers Ready?

As always, every regulatory model and policy has direct implications for conceivable business models, whatever the other policy considerations.

Consumer Internet access, presently mandated as “best effort only,” allows no creation of quality-assured access services that might have direct end user experience benefits for consumers.

Voice and real-time video communications are among the historic examples, as both are sensitive to latency, the problem packet prioritization tries to solve.

Inability to support quality mechanisms for consumer Internet access has been a business model issue for mobile and fixed Internet service provider services.

That is not true for business-focused public hotspot services.

An emerging issue is the degree to which quality-assured services will become a bigger potential business for networks built upon consumer platforms.

Business services are allowed to use such quality of service mechanisms without restriction.

And that is one angle Comcast and others are counting upon to create new revenue streams from a network that has largely, to this point, been an important Internet access feature, but not much of a direct revenue generator.

Comcast and others hope that will change as network footprint increases, seamless authentication and handoff to mobile networks (NGH and Passpoint) are possible, and there are extensive roaming agreements, improved quality of service and security.

What then will be possible are business models that in the past have been difficult or impossible.

Quality-assured access will open up new possibilities for business-class services requiring more security and quality of service mechanisms; services based on ultra-reliable machine-to-machine communications, or video entertainment.

It is reasonable to assume that the customers for such services will be other enterprises, including mobile service providers, app providers and possibly device manufacturers, to an extent.

Since the likely application includes carrier voice and messaging or other business services, there are no network neutrality infractions.

Given the growing importance of traffic offload from mobile networks to Wi-Fi, the likelihood is that both retail and wholesale opportunities to support offload of carrier voice and messaging are possible, creating a new wholesale opportunity for Xfinity Wi-Fi, for example.

source: Wireless Broadband Alliance

Comcast’s 11 million public-facing  hotspots have achieved at least one objective so far: access is surprisingly widespread. I sometimes get signal in the middle of Denver intersections with six lanes one way, four lanes the other.

The network is not intended to substitute for the mobile network, but it is surprisingly useful as an untethered access mechanism many of the places I move about, when in a Comcast region. I haven’t found it works quite as well when outside a Comcast region, but in a zone where Cable Wi-Fi roaming agreements are in place.

Though it is hard to see AT&T or Verizon sourcing capacity from Comcast to densify coverage, others might be more willing, and that is one of the potential enterprise revenue streams Comcast always as aimed for, in building the Wi-Fi hotspot network.

"Level Playing Field" Impossible, or Only a Matter of Policy?

One of the least-controversial statements one can make about competition in any market is that a “level playing field” where all contestants are treated alike is a good policy. The problem always comes in the detail. In most markets, contestants are not literally treated the same.

In markets where wholesale access is the main way competition is enabled, the underlying facilities provider is treated quite differently than everyone else.

In markets where mandatory wholesale, provided by one provider, is not the only form of competition, incumbents are governed by rules that do not apply to any other contestants.

That typically is the case for former telecom incumbents, which almost universally face obligations no other contestant faces. In a few markets (North America and Singapore perhaps illustrative in that sense), there are multiple facilities-based providers. Even there, the playing field is not level.

There tends to be a former incumbent operation operating with wholesale obligations, and then one or more facilities-based competitors who do not have any mandatory wholesale obligations.

And that raises regulatory issues. At what point, if ever, does the playing field really need to be harmonized? When can all contestants--at least facilities-based contestants--play by the same rules?

Presumably there still will be differences, even if all facilities-based providers were to be governed by precisely the same rules. Wholesale operations presumably would not have the obligations the facilities firms face.

It is a difficult issue. In Singapore, facilities-based competition almost makes robust wholesale unnecessary. U.S. regulators, while maintaining most incumbent obligations, opted for an approach that creates incentives for facilities-based competition nearly two decades ago.

But anyway one looks at the matter, treating everyone the same is not the practice, anywhere.

Project Loon Faces Interference in India, for the Moment

Who gets to use spectrum, and concerns about interference from other users, now appears to be an issue for Google’s Project Loon in India.

Project Loon proposes the use of the 700 MHz to 900 MHz spectrum for its downlinks and uplinks, but those frequencies are allocated for mobile phone service.

A concern about potential surveillance is a concern of the Defense ministry.

But the chief obstacle is the objection by the Department of Telecommunications, which notes signal interference could be an issue.

This is not the sort of issue Project Loon would have missed anticipating, especially since Project Loon presently views itself as a backhaul technology working with retail mobile service providers.

In Indonesia, for example, the big mobile companies are partnering with Project Loon for a test of the service, and that is not something they would have done if it was believed there is no way to avoid interference.

In2016, the top three mobile network operators in Indonesia--Indosat, Telkomsel, and XL Axiata--will test Project Loon-- balloons to deliver Long Term Evolution 4G signals across the country.

Sri Lanka might be going further, and reportedly is planning on relying on Project Loon to cover the island for Internet access, starting as early as March 2016. There would seem to be some significant issues to be settled, if that timetable is to be kept.

Unless something new has been developed, Project Loon would beam signals to stationary antennas, much as a fixed wireless network would do. There would be no way to deliver signals directly to a mobile phone using LTE, for example.

So though mobile operators would be logical partners, they are not the only logical partners. Fixed network telcos might arguably be better positioned to serve as on-the-ground sales, installation and support partners.

There also has been confusion about whether the service was to be supplied “for free, or for fee.” Both could be correct conclusions, though it seems likely any “free” services would be limited, both in terms of usage or speed. Think of a program similar to Wi-Fi at public institutions.

It is possible, perhaps likely, that some amount of basic service could be made available “for free.” But it seems highly unlikely that most of the service can be provided that way. Nor would most offers of that sort provide unlimited access or large buckets of usage.

But it seems most likely that most of the service will be supplied at some commercial rate.