1. Network Aware Optimization Technology

In today's mobile networks, laptops, netbooks, and smartphones generate the vast majority of data traffic. These high end devices account for more than 90% of overall mobile broadband traffic.

In this article, we introduce the idea of network aware optimization, and explore how this can benefit mobile operators in addressing these challenges...

(Keep Reading...)
Download Mobixell Network Aware White Paper

2. Smart Charging - It doesn't all have to end in (volume) Tiers?

When AT&T first introduced the iPhone, the "all you can eat" data package seemed to be a natural use of the then under-used mobile data infrastructure. The iPhone's success, followed by the exponential growth in mobile data, has been lengthily discussed. Some say it is a classic "be careful what you wish for" syndrome.

AT&T's move to tiered data plans last June, sounded the first whistle for other operators to follow. But are volume-based tiered packages the best way to harness this data growth?

(Keep Reading...)

3. Mobixell Introduces Load Distribution Management (LDM) to Reduce Total Cost of Ownership (TCO) of Mobile Internet Infrastructure and Traffic Management

When data traffic grows, so do the traffic management solutions and the IT network infrastructure to support them. One of the essential network components required in most redundant implementations is the role of the Network Load Balancers which are in charge of distributing traffic between the different servers and share the load in case one of the nodes has failed.

The modern network load balancers have evolved to include multiple capabilities as well, and as such, also play a very significant role in the Total Cost of Ownership of Traffic Management solutions.

Mobixell identified the need to reduce the system's costs and has therefore developed and launched the Seamless Access Load Distribution Management (LDM). The LDM is a converged mobile data IP traffic balancing software solution that provides mobile operators with the balancing, availability and intelligent traffic management capabilities of a traditional hardware-based load balancer, lowering the Total Cost of Ownership (TCO) by as much as 65-75 percent.

(Keep Reading...)

4. Mobile Data Optimization as an Offloading Technique

As data volumes continue to grow, mobile network operators are faced with the challenge of how to manage the cost of handling this traffic, the majority of which is for over-the-top Internet services (rather than MNO host services). Data offloading is being presented as one technique for addressing this challenge. The question that comes to mind, however, is: Should mobile data optimization also be considered an offloading technique?

(Keep Reading...)

5. Questions and Answers: an interview with Noam Green, Mobixell's AVP of Marketing

Noam Green discusses controversial data growth reports published by industry analysts; provides an interesting view on how operators can offer advanced premium service creation such as "Child Safety"; and elaborates on Device-Aware Optimization techniques and why are they important

(Keep Reading...)

6. Market Insight - Will Mobile Operators be able to change the business model and charge service providers to pay fees linked to usage?

The shift from operator-controlled "walled garden" services, to over-the-top services and the surge in data usage, has caused some mobile operators to fear that they are investing millions in their infrastructure for others to reap.

Can MNOs change the rules of the game?

(Keep Reading...)

Data optimization techniques have been used for many years in mobile networks to improve the user experience when using mobile data services. It can be argued that the early WAP1 protocol was introduced not only to deal with the limited capabilities of early devices but also to optimize the delivery of content over limited bandwidth circuit switched data connections. However, it was the introduction of 2.5G data cards enabling laptops to access the open Internet across the mobile network that was the catalyst for the introduction of Web optimization platforms into the mobile data network.

Mobixell's data optimization includes both Web optimization/acceleration, as well as video optimization. While Mobixell's video optimization uses sophisticated state of art techniques which can reduce the volume of video by 40% with no visible degradation in user experience, even when using state-of-the-art compressed domain transrating techniques, the cost of optimizing a video is significant when compared to just forwarding packets through the mobile network.

When one considers that during the busy hours, only 7% of the cells are congested (source: Vodafone Group PLC - Interim Management Statement Feb 2010) the case for being more intelligent about how and when to optimize video becomes clearer. Mobixell's approach is that of network aware optimization via which, rather than taking a brute force "optimize all" approach, a holistic view is taken to select which videos to optimize.

The first part of this holistic view is to look at the quality of the data connection to the user and determine if this is sufficient to allow the user an uninterrupted viewing experience. If it is not, then Mobixell's Seamless Access solution can reduce the bit rate of the video to that which allows the user to play the video without stalling. This bandwidth detection is continually measured and the rate of optimization adjusted during the video playback.

After looking at individual users, Seamless Access takes an aggregated view of the mobile cell to determine if a cell capacity threshold is being approached. Once this capacity threshold is reached, Seamless Access will start to optimize video in order to reduce volume of data in the cell. This is done in order to try and keep the cell from becoming congested.

Being network-aware is not just useful for video optimization, but can also be used to improve the user experience by targeted Web optimization. Many MNOs measure their CTOs by the elapsed time taken to load certain Web pages. Web optimization platforms have now gone beyond volume reduction techniques, and employ techniques to reduce the page load times (also referred to as Web acceleration). Some of these techniques involve tricking the browser into issuing more concurrent requests. If these techniques are used ubiquitously then they can have negative impact on congested cells. Being network-aware allows the correct Web acceleration techniques to be used, given the dynamically changing set of network conditions.

Network-aware optimization allows MNOs to balance investment against capacity management and improved user experience. Brute force optimization cannot be used to manage the surging data traffic. By implementing intelligent optimization techniques, MNOs can protect their networks from the surging data network, while at the same time enhance their users' perception of their network speed and efficiency.

Back to Top

Typical subscribers do not pay attention to what data volumes they consume. If I asked each of you how much volume you consumed on the Internet last month, the vast majority of you would have no idea. But, if I asked you how you utilized the Internet, you would be much more likely to know the answer - e-mail, YouTube, iTunes, Facebook, etc. You would have a pretty good idea of the services and applications that you use. These mean something to you, much more so than the number of gigabits does.

Volume tiers can be confusing

As a result, we will see tiered volume data plans become popular for a time, but in the long run, there will be a flat-rate fee subscribers will pay along with other "usage-related" fees buried in the cost of other services. Users may be more inclined to pay for different application-level packages they can relate to (and quantify), such as flat-rate browsing with a two-hour video viewing limit, or social networking packages. These easily quantifiable tiers would remove some of the resentment users have voiced lately regarding the change to tiered pricing. These users feel they were penalized for using their data-hungry devices. Operators also have an opportunity to monetize on the over-the-top data by bundling the application price and its corresponding data. For example, when you download a new book to your Kindle, how much does that data volume cost you on the mobile network? The answer is you have no idea; it is bundled in the price of the e-book that you downloaded. Subscribers accept that. They understand the cost of a book. It fits their paradigm. This bundling mechanism can work for both the service provider, as well as the mobile operator.

One important factor should be kept in mind when offering differentiated service offering, and that is the technology behind the scenes which needs to support the offering. For example, in order to enable users to browse freely on one site (e.g: Facebook) and pay for browsing on a different site (e.g: YouTube), a technique called Zero Rating needs to be implemented. Zero Rating means the operator can employ a free-of-charge rating when consuming a specific service. As an example, a mobile operator can provide a free-of-charge browsing experience on facebook as part of his "social networking package". While it may seem simple to configure the GGSN (or DPI) to "not charge" for any URL coming from Facebook, it requires a bit more technology to do this right. If you take a look at your standard Facebook page, you may see videos, applications, ads, and other embedded content. What seems to be coming from Facebook only, is actually a mixture of multiple sources (YouTube, ad networks, images from other pages, etc.). In order to provide true Zero Rating, a technique of analyzing the session is needed to enable to connect between facebook-embedded YouTube videos vs. YouTube videos embedded in other sites.

Operators are only now starting to realize the potential of what we at Mobixell refer to as "Smart Charging". Operators would, therefore, be well advised not to conclude that tiered data plans will be the solution to their economic challenges, but rather just a piece of the puzzle. More essential to their long-term prosperity, is their ability to act upon mobile data streams at a granular level. In so concluding, mobile operators should prioritize intelligent, flexible decision-making as a fundamental requirement for their future mobile Internet and mobile broadband infrastructure platforms.

Back to Top

It is forecast that mobile operators will be increasing their network capacity at a faster rate than ever before over the coming years, an unprecedented 108 percent compound annual growth rate (CAGR) to 20142. Software scalability in the traditional hardware approach is often limited by the network infrastructure in which it was initially deployed. That means network infrastructure, especially around the load balancer component, has to be replaced as often as every two years.

In order to reduce the Total Cost of Ownership of Mobixell's Seamless Access solution, Mobixell has developed Load Distribution Management (LDM), a converged mobile data IP traffic-balancing software solution that provides mobile operators with the balancing, availability and intelligent traffic management capabilities of a traditional hardware-based load balancer.

The new Seamless Access module lowers the Total Cost of Ownership (TCO) of network upgrades by as much as 65-75 percent, while providing increased flexibility in network planning and implementation along, through incremental, de-facto unlimited expansion of network capacity and total bandwidth.

Seamless Access LDM lowers the TCO of network capacity upgrades by enabling mobile operators to eliminate the usual need to overhaul hardware-based load balancers (performing a 'forklift upgrade') but perform incremental expansion of the network capacity in line with the actual bandwidth requirements. With this new module, the Seamless Access, fundamentals of incremental, unlimited, growth also apply to the load-balancing stage of today's fast growing mobile data access solutions. The module achieves the same results as a hardware-based solution with regards to load balancing, health detection, and service availability through an innovative combination of server resident software and Policy Based Routers (PBR).

The Seamless Access LDM is already in production in multiple Tier-one mobile network operators in Europe, Latin America and Asia Pacific.

Back to Top

Nearly every presentation at a the recent Broadband Traffic Management Conference used the now ubiquitous hockey stick graph illustrating the diverging data volume and data revenues projections, and many presenters discussed ways for the mobile operators to manage this issue and maintain the profitability of their data services. In addition to the topics of variable charging and traffic management, for example via QoS and bandwidth shaping, part of the conference focused on offloading as a technique for helping the MNOs manage the cost side of providing data services.

Wifi and Femto offload in addition to improving coverage in poor coverage areas can be used as techniques to offload the radio network and the backhaul networks. The 3GPP via SIPTO (Selective IP Traffic Offload) is standardizing the technique offloading the core network by allowing selective traffic to be sent directly to the Internet rather than through the operator's core network. These offload techniques help the MNOs to manage their investment and reduce their costs of providing a data service.

Mobile data optimization - especially video optimization - provides many of the same attributes of these offloading techniques. The majority of the traffic in mobile data networks is downstream and a significant portion of this is video, which has been projected to be upwards of 60% of all data volume in the future. Mobixell's Seamless Access Video Optimization can reduce the volume of video by approximately 40% while maintaining the user's quality of experience. This reduction saves on RAN and backhaul capacity, as well as the volume of data flowing through the core network, video caching also reduces the volume of data at transit points.

Mobile data optimization is an effective way of offloading the data network and can be deployed quickly and effectively in the core network, using targeted network-aware optimization as provided by Mobixell's Seamless Access further improves the ROI of deploying mobile data optimization.

This point of mobile data optimization being an effective offloading technique, was re-enforced by Aditya Kaul of ABI Research during the Broadband Traffic Management Congress in London. The following slide from his presentation shows the volume of traffic covered by the different offloading techniques 2015 and illustrates the impact of mobile data optimization as an offloading technique.

ABI Research: Offload Techniques 2015

Back to Top

Q: Analysys Mason published a recent report noting data market growth in Europe is over-hyped. What's your take on this?

A: The data growth numbers exposed in Analysys Mason's report are intriguing, and our installations around Europe have indeed indicated a slowdown in the exponential growth shown in previous years. But, contrary to the report, our numbers collected from the field show significant growth of about 60% in about six months in a Tier 1 Southern European mobile operator. Other Western European Tier 1 operators have shown lower yearly growth rates of about 40%-50%. While the report concentrates on the European market, it is important to note that the situation in South East Asia seems to be somewhat different, showing growth rates almost quadrupling in the past year.

Mobixell's collected data continues to support indications that mobile video is continuing to lead the highest growth in Internet traffic, and operators will still need to look at innovative solutions to save costs. As mobile video continues to improve in quality (most recently with HD quality video), the size of video files also increases, emphasizing that effective mobile video optimization is the way to provide for future mobile users, whatever the network data capacity.

Q: There is a lot of discussion around how to reduce the traffic load and reduce TCO, but what do you see as revenue generating sources for operators in this market?

A: It's no secret that mobile Internet and over-the-top services are taking a bigger share of user's attention (and dollars). Nevertheless, I believe mobile operators can still play an important role in providing premium services mainly around the areas of content filtering and content security.

An example of an innovative service recently offered by Mobixell is what we call "Child Safety" which is essentially the next generation Parental Control. One of the main problems around existing parental control solutions are that they are in actual fact URL filtering mechanisms - They allow or prevent users to access specific sites. While controlling access to adult sites is an important part of Parental Control, these sites are only a small part of youngsters' mobile activity.

One of the main activity youngsters are busy with on their mobile devices is messaging: SMS, MMS, IM, social networking site messaging etc. All these channels flow through the operator, and thus offer a unique place to monitor and control them.

Mobixell has partnered with a leading parental control supplier to provide Conversation-Aware parental control in order to protect the youngsters from all kinds of online threats including Sexting, Cyberbullying and Online Predators. The solution provides parents the ability to monitor or intervene with potentially damaging conversations. Realtime SMS and image recognition techniques allow parents to keep "in the know", something that parents are missing sometimes in this fast-paced generation.

Q: According to ABI Research, smartphones made up 19% of all handsets shipped in the second quarter of 2010. How does this fact affect mobile data management and optimization solutions?

A: Smartphones seem to be living up to their predictions to be the next moving force of the mobile data Internet, currently led (volume-wise) by dongles and laptops. But even the generic word - smartphones - is highly segmented. As we know, Apple devices and Android devices have different OSes, different App Stores, and different video playback capabilities. And even then, the different devices within each product set have almost as many differences between them - for example, the iPad is not just a big iPhone. There are multiple differences which need to be considered other than the obvious screen size: CPU power, browser and video player capabilities, etc.

Legacy optimization solutions tend to treat and optimize all data equally, if it is directed to a PC or a smartphone which essentially leads to making use of the lowest common denominator between different devices. This contributes to lower quality of service. In order to retain the best quality of service, optimization solutions need to include device-aware capabilities to provide the right optimization techniques according to the target device. As an example, one would think that the video capabilities of the iPhone and iPad are identical. Our tests have shown that the iPad's stronger computing power enables it to use a higher compression video codec than the iPhone 3GS (H.264 high profile vs. H.264 base profile). The differences between these codecs mean that the operator can provide a higher quality video experience for iPads using the same compression rate as that used for the iPhone. While each operator has a different smartphone "blend" (iPad/ iPhone4/ iPhone3/Androids, etc.), implementing Device-Aware capabilities can enable about 30% of data savings while providing an almost identical video quality.

The diversity in the smartphone domain as well as the lack of information regarding the limitations of each device has led Mobixell to make use of its manually-tested device database to enable device-aware optimization in its products.

Back to Top

Mobile operators have been investing billions of dollars in network investments needed to support the surging data traffic, and especially bandwidth-hogging services such as video. The problem is that while the critical volume metrics have all grown substantially, they are not growing on the same trajectory - specifically, the growth of mobile data volumes are beginning to outstrip the growth in revenues, and importantly a mobile operator's cost structure is defined more by mobile data volumes than by the number of subscribers.

The growing gap between operator's revenues and traffic growth

Mobile operators have been struggling with how to address this issue, and it is only recently, that MNO telecom executives such as France Telecom, Telecom Italia, and Vodafone Group, have started to voice their views on the subject, trying to pass some of the costs on to the service providers.

In last December's "Le Web" conference in Paris, Stephane Richard, France Telecom CEO, mentioned that this mismatch between investment and revenue "is set to compromise the economic sustainability of the current business model for telecom companies".

Cesar Alierta, Telefonica's CEO, was quoted last February saying that companies such as Google and Yahoo! "... use Telefonica's network for free, which is good news for them and a tragedy for us"... "That can't continue".

Service providers such as Dailymotion have been quoted in Bloomberg's December article saying they already pay enough. "Currently about 40 percent of our expenses go to networks anyway -- servers, peering, our content delivery network, and other resources," said Giuseppe de Martino, the legal and regulatory director of Paris-based online-video provider Dailymotion SA. "If telecom operators want us to share in their expenses, perhaps we should talk about sharing subscription revenues as well."

Can mobile operators gain enough traction to change the rules of the game? Can they convince net-neutrality regulators to enable operators to charge consumers or content providers a premium for delivery of certain videos, games, or other applications?

Whether operators like it or not, there are other significant players that will have their say.

Back to Top