Why are the US government and DoD in particular interested in Open RAN?

Over the last 24 months, it has been very interesting to see that the US Government has been moving from keen interest in Open RAN to make it policy for its procurement of connectivity technology.

As I am preparing to present for next week's RIC Forum, organized by NTIA and the US Department of Defense, many of my clients have been asking why the US Government seems so invested in Open RAN.

Supply chain diversification:

The first reason for this interest is the observation that the pool of network equipment provider has been growing increasingly shallow. The race from 3G to 4G to 5G has required vendors to attain a high level of industrialization and economy of scale, that has been achieved through many rounds of concentration. A limited supply chain with few vendors per category represents a strategic risk for the actor relying on this supply chain to operate economically. Open RAN allows the emergence of new vendors in specific categories that do not necessitate the industrial capacity to be delivering end to end RAN networks.

Cost effectiveness:

The lack of vendor choice has shifted negotiating power from network operators to vendors, which has negatively impacted margins and capacity to make changes. The emergence of new Open RAN vendors puts pressure on incumbents and traditional vendors to reduce their margins.

Geostrategic interest:

The growth of Huawei, ZTE and other Chinese vendors, with their suspected links to the Chinese government and Army, together with the somewhat obscure privacy and security laws there, has prompted the US government and many allies to ban or severely restraint the categories of Telecom Products that can be deployed in many telecom networks.

Furthermore, while US companies dominate traffic management, routing, data centers and hyperscalers space, the RAN, core network and general telco infrastructure remains dominated by European and Asian vendors. Open RAN has been an instrument to facilitate and accelerate Chinese vendors replacement, but also to stimulate the US vendors to emerge and grow.

DoD use case example: Spectrum Dominance

This area is less well understood and recognized but is an integral part of US Government generally and DoD's in particular interest in Open RAN. Private networks require connectivity products adapted for specific use cases, devices and geographies. Commercial macro networks offer "one size fits all" solution that are difficult and costly to adapt for that purpose. Essentially DoD runs hundreds of private networks, whether on its bases, its carriers or in ad hoc tactical environments. Being able to setup a secure, programmable, cost effective network, either permanently or ad hoc is an essential requirement, and can also become a differentiator or a force multiplier. A tactical unit deploying an ad hoc network might look at means not only to create a secure subnet, but also to establish spectrum dominance by manipulating waveforms and effectively interfering with adverse networks. This is one example where programmability at the RAN level can turn into an asset for battlefield dominance. There are many more use cases, but their classification might not enable us to publicly comment them. They illustrate though how technological dominance can extend to every aspect of telecom.

Open RAN in that respect provides programmability, cost effectiveness and modularity to create fit for purpose connectivity experiences in a multi vendor environment.

What are today's options to deploy a telco cloud?

Over the last 7 years, we have seen leading telcos embracing cloud technology as a mean to create an elastic, automated, resilient and cost effective network fabric. There has many different paths and options from a technological, cultural and commercial perspective.

Typically, there are 4 categories of solutions telcos have been exploring:

• Open source-based implementation, augmented by internal work
• Open source-based implementation, augmented by traditional vendor
• IT / traditional vendor semi proprietary solution
• Cloud provider solution

The jury is still out as to which option will prevail, as they all have seen growing pains and setbacks.

Here is a quick cheat sheet of some possibilities, based on your priorities:

Obviously, this table changes quite often, based on progress and announcements of the various players, but it can come handy if you want to evaluate, at high level, what are some of the options and pros / cons of deploying one vendor or open source project vs another.

Details, comments are part of my workshops and report on telco edge and hybrid cloud networks.

Announcing telco edge computing and hybrid cloud report 2020

As I am ramping up towards the release of my latest report on telco edge computing and hybrid cloud, I will be releasing some previews. Please contact me privately for availability date, price and conditions.

In the 5 years since I published my first report on the edge computing market, it has evolved from an obscure niche to a trendy buzzword. What originally started as a mobile-only technology, has evolved into a complex field, with applications in IT, telco, industry and clouds. While I have been working on the subject for 6 years, first as an analyst, then as a developer and network operator at Telefonica, I have noticed that the industry’s perception of the space has polarized drastically with each passing year.

The idea that telecom operators could deploy and use a decentralized computing fabric throughout their radio access has been largely swept aside and replaced by the inexorable advances in cloud computing, showing a capacity to abstract decentralized computing capacity into a coherent, easy to program and consume data center as a service model.

As often, there are widely diverging views on the likely evolution of this model:

The telco centric view

Edge computing is a natural evolution of telco networks. 
5G necessitates robust fibre based backhaul transport.With the deployment of fibre, it is imperative that the old copper commuting centers (the central offices) convert towards multi-purposes mini data centers. These are easier and less expensive to maintain than their traditional counterpart and offer interesting opportunities to monetize unused capacity.

5G will see a new generation of technology providers that will deploy cloud native software-defined functions that will help deploy and manage computing capabilities all the way to the fixed and radio access network.

Low-risk internal use cases such as CDN, caching, local breakout, private networks, parental control, DDOS detection and isolation, are enough to justify investment and deployment. The infrastructure, once deployed, opens the door to more sophisticated use cases and business models such as low latency compute as a service, or wholesale high performance localized compute that will extend the traditional cloud models and services to a new era of telco digital revenues.

Operators have long run decentralized networks, unlike cloud providers who favour federated centralized networks, and that experience will be invaluable to administer and orchestrate thousands of mini centers.

Operators will be able to reintegrate the cloud value chain through edge computing, their right-to-play underpinned by the control and capacity to program the last mile connectivity and the fact that they will not be over invested by traditional public clouds in number and capillarity of data centers in their geography (outside of the US).

With its long-standing track record of creating interoperable decentralized networks, the telco community will create a set of unifying standards that will make possible the implementation of an abstraction layer across all telco to sell edge computing services irrespectively of network or geography.

Telco networks are managed networks, unlike the internet, they can offer a programmable and guaranteed quality of service. Together with 5G evolution such as network slicing, operators will be able to offer tailored computing services, with guaranteed speed, volume, latency. These network services will be key to the next generation of digital and connectivity services that will enable autonomous vehicles, collaborating robots, augmented reality and pervasive AI assisted systems.

The cloud centric view:

Edge computing, as it turns out is less about connectivity than cloud, unless you are able to weave-in a programmable connectivity. 

Many operators have struggled with the creation and deployment of a telco cloud, for their own internal purposes or to resell cloud services to their customers. I don’t know of any operator who has one that is fully functional, serving a large proportion of their traffic or customers, and is anywhere as elastic, economic, scalable and easy to use as a public cloud.

So, while the telco industry has been busy trying to develop a telco edge compute infrastructure, virtualization layer and platform, the cloud providers have just started developing decentralized mini data centers for deployment in telco networks.

In 2020, the battle to decide whether edge computing is more about telco or about cloud is likely already finished, even if many operators and vendors are just arming themselves now.

Edge computing, to be a viable infrastructure-based service that operators can resell to their customers needs a platform, that allows third party to discover, view, reserve and consume it on a global scale, not operator per operator, country per country, and it looks like the telco community is ill-equipped for a fast deployment of that nature.

Whether you favour one side or the other of that argument, the public announcements in that space of AT&T, Amazon Web Services, Deutsche Telekom, Google, Microsoft, Telefonica, Vapour.io and Verizon – to name a few –will likely convince you that edge computing is about to become a reality.

This report analyses the different definitions and flavours of edge computing, the predominant use cases and the position and trajectory of the main telco operators, equipment manufacturers and cloud providers.

2013: the year of Big Video

In September of 2012, Vodafone Germany shocked the industry in announcing that video was consuming 85% of its LTE network.
As we have started a brand new year and vendors and operators alike are finalizing plans for Mobile World Congress, I thought it would be timely to review what the main vendors of video optimization were up to in 2012.

Allot / Ortiva Wireless:
The company, with the acquisition of Ortiva Wireless in April and Oversi in July has certainly made great strides in their strategic plan to provide a one-stop-shop traffic management solution to its customers. Between DPI, policy management, charging functions, transparent caching and video optimization for mobile, enterprise and fixed broadband, the company has a large tool set and addressable market. The challenge will be in the integration of the acquired technologies and talents, together with the formulation of a differentiating, competitive solutions offering that goes beyond analytics, charging, managing...

Avvasi:
The company has entered the fray with a fresh outlook. Leader in mobile video QoE measurement, they have been asked by their customers to help manage the video QoE and have launched a new product (Q-SRV) to "measure, manage and monetize" the video experience in mobile networks. One to watch in 2013.

Bytemobile / Citrix: 
The acquisition of Bytemobile by Citrix last year was a big shock for the market segment. The leading vendor in market share was acquired by an industry's outsider under the rationale to enter the mobile market. Citrix is definitely starting to tickle F5 and Cisco with Netscaler as a load balancer / proxy in mobile networks. What best introduction in the mobile space than the leader in mobile internet and video optimization? It will be interesting to watch how the replacement of Unison platform by the T3000 series, together with the suggested replacement of F5 by Netscaler plays out in the coming months...

Flash Networks:
The company started the year with a bang with 3 large tier 1 customers in video optimization (MTS russia, Globe telecom and Telekom Austria Group) announcements. Since then, it has been quiet, but the company has been busy upgrading and up selling their existing customer base. It will be interesting to see whether the company will take advantage of some of the market deals coming this year.

Huawei:
The secretive Chinese vendor has seen its market share in browsing gateway increase dramatically over the last 3 years with over 114 operators and this is the traditional Trojan horse for video optimization to enter the market. Mobile internet is becoming a focus on the company's strategy in the core network and it looks like the slides on video optimization presented last year in Barcelona are morphing into a product offering. It is still early days, but Huawei can move fast when needed.

Mobixell Networks:
The company has digested its 724 solutions acquisition and made good progress in converting its installed base and winning new deals in video optimization. Not enough to satisfy its investors, apparently, with the replacement of its CEO in December last year. The company is clearly looking for different growth parameters and it is likely that we will see more strategic activities from them in 2013.

NSN:
Not much to report for the Nordic giant, struggling to impose its vision of self organizing networks in core. The sale of its OSS/BSS division to Redknee will either see a refocusing or spin off of the video assets.

OnMobile:
The company acquired Dilithium assets in 2010 and has been struggling to have a video strategy since. They briefly considered video optimization, but video is no longer a focus for the company in 2013.

Openwave Mobility:
After a rough couple of years, seeing its CEO replacement, and the spin off to an equity venture, the company is starting to re focus and to reboot its traffic management and video optimization strategy. They have a few references in the space and are working to update their browsing gateway's installed base with video optimization. It will be interesting to see if they are up to the challenge in 2013.

Skyfire:
The company took the market by storm by launching cloud-based video optimization and simultaneously announcing Verizon Wireless as a customer and investor. Since then, with a fresh round of financing, the company has been expanding its reach to Europe. It will be good to see how cloud takes in mobile networks.

In September of 2012, Jens Schulte-Bockum , CEO Vodafone Germany shocked the industry in announcing that the 10% of their customer base who have elected to shift to their LTE network had a fundamentally different usage pattern than their 3G counterparts:

“Voice, text, other messaging and data - everything that makes money for us - uses less than 15%. The bit that doesn’t make money uses 85% of the capacity. Clearly we are thinking about how we can monetise that. ”

“The bit that does not make money for us” is mobile OTT video. As mobile video threatens to overgrow every other traffic types, operators start to look at ways to alleviate the costs associated with the necessary capacity upgrades to meet the demand, as well as strategies to monetize this large, untapped opportunity. I will be releasing my report "Mobile Video Optimization 2013" on March 15 where I will examine the latest strategies from the dominant vendors in the space.

Video optimization 2.0, market reset

On the heels of broadband traffic management's show in London last week, I thought it was time to take stock of that market segment as most vendors have launched their second generation product recently.

The market leader, Bytemobile (with 55% market share of deployments), started the trend this summer, when launching their new dedicated appliance, the T-3000. While this is not strictly a new version of their Unison product, it is a new computing platform sold as an appliance, departing from the software infrastructure business model. It is a first step towards solving some of the scalability issues experienced by the former solution, that saw dpi, policy, charging, web and video optimization inextricably amalgamated, whether you wanted to use all products or not. It gets rid as well of these expensive load balancers that were a high cost low yield proposition. Bytemobile is not the only one to experience price pressure and to take the knife to load balancing as the bandwidth requirements increase.

Mobixell, with 16% market share, seems to be at last in a position to digest their 724 solutions acquisition. While both product lines were quite complementary and had little overlap, it was a tough proposition for Mobixell to acquire 724, rationalize the technologies and workforce and face the ire of their traditional resellers and OEM (NSN, Huawei, Ericsson...). These were weary to see their supplier compete head to head with them in mobile broadband as Mobixell was rolling out 724 seamless gateway proposition along with their streaming and transcoding platform. The result saw Mobixell practice a tough price attrition in the market, helped by a low cost structure (724 solutions technology comes with integrated routing and inter process UDP-based communication that provides great scalability at low cost). Mobixell announced the launch of the new product release, called EVO, taking some of the computational power to the cloud. While some are skeptical about how much can be accomplished in the cloud for real time video optimization, it certainly is a good step towards cost and CAPEX containment worth exploring.

Flash networks with 8% has been quite busy on the market, silently plowing ahead, upgrading existing customers and winning a handful of deals. They have announced the new version of their product and are as well taking a big step in technology investment in that space.

Ortiva wireless with 3% market share has seen some very good progress this year, bagging some good high profile accounts, nearly tripling their year on year revenue, from an admittedly small footprint. The company has not announced a new version of their product yet, staying on their existing appliance model.

Skyfire labs, with 2% market share, a very innovative start up with a cloud based approach, evolved from their tablet and smartphone browsing app has also been able to grab some high profile tier 1 carrier, together with high profile VAR agreement with infrastructure vendors.

Openwave, with 1% market share, as you know, has had a very busy year on the corporate and financial front (here, here, here, and here), but has not announced much from a product, technology or customer standpoint. They are fighting for their survival and seem to be focusing to a return to financial stability (PS revenue increase, licensing of their patents to Microsoft) before investing further in technology or customer acquisition.

NSN has been developing their homegrown technology, wanting to end the reliance on their traditional partners in the space and came out with a very basic first attempt, focused around loss-less transmission. Nowadays, they are trying to push their "liquid" network concept and seem to be going at it in a fairly scattered manner.

These new product announcements signal, beyond the usual technology investment from start ups and established vendors, a market reaching a level of maturity fast, only 2 years after inception. Some might even say that this segment is commoditized before having really taken off. According to my calculations, this is a market that has generated about $90 millions for vendors this year. We can see from the number of players why price attrition plays an important role, even though traffic is increasing fast. We will see some consolidation and attrition in that space soon, as insufficiently capitalized vendors wont be able to sustain the market growth.

RGB networks, Juniper, Cisco, Huawei, Acision are all active in this space too, while others are preparing to enter the market. The market share are {Core Analysis} calculations, part of an upcoming report on the mobile video optimization space. Details and questions can be addressed here or at patrick.lopez@coreanalysis.ca.

Openwave CEO replaced - Consolidations to come in the traffic management market

Openwave announced today the resignation of its CEO, Ken Denman, quoting personal reasons. Denman is being replaced by Anne Brennan, the company's CFO.

As we have seen in a previous post, Openwave has been struggling for a while to deliver on the expectations it has raised in the market to provide an integrated traffic management solution for video.

After failing to show the results on over 40 announced trials, after failing to upsell their installed base with their next generation of products, after after buying back old patents and suing RIM and Apple, Openwave sees its CEO resign and, the same day,  is nominating Peter Feld as Chairman of the Board, replacing Charles E. Levine.

This market segment, born from the ashes of the wap gateway market, sees companies like Acision, Bytemobile, Comverse,  Ericsson, Flash Networks, Huawei, Mobixell, Nokia Siemens Networks, and others become the intelligent gateway in the network. That gateway's role is to complement and orchestrate DPI, charging, PCRF, video optimization. It is a key network function.

As most data traffic is browsing related, companies that used to sell wap gateway are the best positioned to capitalize on upselling a richer, more sophisticated gateway that can provide means for operators to control, monetize and optimize browsing and video traffic in their network.

Openwave has not been able to negotiate that trend early enough to avoid its market share being eaten up by traditional competitors and new entrants. Additionally, as the traffic has fundamentally changed since tablets and smartphones have entered the market, key capabilities such as TCP, web and video optimization were late to appear in Openwave's roadmap and proved challenging to build rather than buy.

Mobixell started the consolidation with the acquisition of 724 solutions last year.
I bet we will see more consolidations soon.

Worst of breed, golden silo

After some 13-odd years in new technology and product introduction, you can't help but look at trends that pop up in this industry with a cynical eye.

One of the catch phrases I hear often is best-of-breed approach. For me, every time I hear it, it is a sure sign that a market segment or a technology is not mature.

It is somewhat counter-intuitive. Best-of-breed pick-and-choose componentized approach to service delivery hints at ranges of well defined components, fungible, interchangeable. You would think that the interfaces being well defined, each vendor competes on unique differentiators, without impacting negatively the service delivery.

Conversely, silo has become an increasingly bad word in telecoms, evoking poorly architected, proprietary daisy chain of components that cannot integrate gracefully in a modern organic network.

Then why is it that best-of-breed always end up taking longer, being more expensive and less reliable than a fully integrated solution from a single vendor?

In my mind, the standards that have been created to describe the ideal networks, from WAP to MMS, from IMS to LTE have been the product of too many vendor lobby-ism. The results in many case are vaguely defined physical and functional components, with lowest common denominator in term of interface and call flows.
The service definition being somewhat excluded from standards has left little in term of best practice to integrate functional components efficiently.

There is a reason in my mind why the Chinese vendors ZTE and Huawei are doing so well. It is not only because of their cost structure, it is because their all in-house technology approach for business critical components make sense.
It allows fast, replicable deployment and trouble shooting. There is much less complexity in integration and roll out, which is the most consuming part of CAPEX.

Whenever you see these vendors using third party technology, it is because it is either so mature and stable that it is not worth developing in-house or so specialized that it has not been developed yet.
In any case, we are talking about fringe technologies. Anything that is business critical is identified long in advance and developed in-house.
Their product and services might not be as sophisticated or differentiating than specialized vendors, but they deliver value by providing the minimum services at the lowest cost, with good enough reliability.

The companies that will win will either be small niche vendors at the periphery of the larger market opportunities or companies that will be good at providing better value, with stronger benefit, but at a price that  is equivalent.