Blog Post

Here Come the Mobile CDNs

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Today, telco gear maker Dilithium Networks launched a software product for carriers, content publishers and content delivery networks that can handle all of the transcoding necessary to take content formatted for one screen and move it to another in real time. The Dilithium Content Adapter is the first software product from the seven-year-old telecommunications gear maker. The company has focused on 3G video since its inception, and Dilithium says the product is already deployed with some operators and CDNs.

But Dilthium’s not alone in its focus on delivering faster video to mobile devices. In a few months, Limelight Networks will launch a mobile CDN product for its customers, and Dave Hatfield, an SVP of marketing sales at Limelight, says customers are testing such a product now. While it’s not a huge focus at Limelight right now, he says phones like the iPhone have changed the potential size of the market by making it easier for consumers to get mobile video — and that could spur market growth.

After the launch of of the iPhone, which opened the Internet to mobile users in ways that were previously cost prohibitive or downright impossible, mobile video may be inching closer to reality. I’m even inclined to shed my doubts about mobile video (although not mobile TV). As such, operators may have to worry about delivering everything from video ringtones to YouTube content on devices. And that could mean a new market for content delivery networks.

Delivering images and video over the Internet to a PC via a CDN is an established fact of doing business for content publishers, but adding mobile screens to the mix have a few gear and service providers seeing green. Such vendors are trying to capitalize on three opportunities in the mobile infrastructure to sell products.

First is some sort of transcoding service, through which content formatted for TVs or PCs is encoded and decoded in real time, or encoded in a variety of formats and stored for delivery to the appropriate device. The second is a sizing service that fits the content to the mobile screen on one of more than 5,000 different mobile devices out there. Finally, the third is any sort of tweak that can reduce the amount of space and time to deliver mobile video on a wireless network.

There are skeptics. Barrett Lyon, CTO of BitGravity, a P2P CDN, scoffs at the notion that any sort of specialized services need to be offered for delivering content to a mobile phone. He points out that CDNs are already delivering ringtones and other content to mobile devices. He may be right, which means Limelight may not find a huge market for its services.

However, I tend to believe that real-time transcoding and other ways of rendering content delivery across multiple devices seamlessly will propel sales of gear or software in the years ahead. Especially if mobile Interent devices take off like chip makers hope.

22 Responses to “Here Come the Mobile CDNs”

  1. Hey did you see that Mobile CDN Inc have launched a new iPhone streaming service with orb for the NBA Allstars games and wiki states that the founder of Mobiel CDN is the guy who was responsible for mobile at Limelight.

    I actually used Limelight for some mobile services and he was a really knowledge base for mobile content. I wonder what this means for Limelight as I don’t see them launching any mobile services.

  2. Hey Arizona man….

    I agree, EDGE sucks! A special mobile CDN is not going to fix the data rate problems that EDGE has… also EDGE is a TCP/IP based function, so any CDN (mobile or not) can feed data to an EDGE device.

    The same applies to EVDO, which has a lot of bandwidth… so to me it’s not the CDN, it’s the data layer. If the data layer is slow, then a CDN no matter how special saucy it is…. will still be slow.


  3. Arizona man


    Wow, this is really political huh? Couple of things to get out of the way. First, while the iPhone is an amazing device, Edge is terrible. There is no way to overcome poor bandwidth. I challenge any CDN to be able to overcome problems faced by a poorly constructed physical delivery system like Edge. That said, it is all TCP/IP, so I get what Mr. Lyon is saying. I think that the larger CDN’s can rename the technology all they want, the issue is cost, and who can deliver the goods at the best cost. Re-naming a product riding on the same expensive network, then seeding a comment thread with negative posts about the only person willing to call it like it is…well it is just hype hype hype…I would love to see a mobile device CDN bake off; comparing delivery, reliability, and price. I have a feeling it would prove Mr. Lyuon correct.

  4. Again… you guys are talking about the wrong stuff. Mobile devices are just little computers, if they are on IP networks then standard CDN technology can delivery bits to these devices without anything special.

    The special stuff is in how the bits delivered to the devices are encoded, it’s all about the content, types of content, and means at which the content is packaged… not about the delivery technology itself.

    It’s pretty clear that marketing for these companies are posting here, which was my point… I could create a “mobile CDN” by just calling myself that, it’s marketing fluff.

    Just think about it for a while.. What is iTunes? It’s a HTTP enabled market place for music. Why would it need to be on a special CDN? Why not just have content which is light that the mobile devices can ingest better than the full blown thing?

    What is T-Zones? It’s a web page… why would it need a special CDN?

    At the end of the day, these are all IP devices that ride on the General Internet. So the same CDNs that deliver to the home desktop can deliver to a phone.

    One may try to argue that the data layer is the special sauce for a mobile CDN, but the data layer is nothing a CDN needs to worry about, if the throughput is there, then IP networks can reach these devices, and it’s all an abstraction.


  5. Ms Higginbotham, thanks for forwarding the URL for the Bitgesvity CTO’s personal page but I would still maintain after reading about his background that he knows very little about mobile.

    Just because he is an expert in his field does not mean he has any knowledge about mobile content delivery.

  6. Franklin

    CDNs enable content providers to deliver better experiences at lower costs.

    Mobile service providers mainly enable mobile networks to deliver better experiences at very high costs that are actully much cheaper than deploying additional cell sites to accommodate new data/ video services.

    If Limelight and Dillithium have the foresight to combine these services at a time when the operators, who control the market, are beginning to open their networks then they have my vote.

    Why does the CTO of Bitgravity have such a problem with this? Does he wish he had read what Dillithium and Limelight had to say before sticking his head into a market he obviously does not understand. Pitty he describes some vision the mobile content industry really needs as ‘marketing fluff’.

    I would reiterate one of the main issues with mobile. That is people who say they understand and make comment about mobile when they have no real knowledge or expertise.

    Great post by the author as it certainly showed the difference between ignorance and insight.

    Note to self, a CTO, think before scoffing on a subject where I need a deeper understanding.

  7. Great article Ms Higgenbotham, who is correct to doubt the viability of any mobile data service as these services has always been a disappointment.

    First, the list provided of Mobile CDN solution providers has one common weakness and is the main problem. None of them actually have a Content Delivery Network (CDN) and hence are more like Mobile Solution providers rather than Mobile CDN providers.

    Most of these mobile solutions are sold to Mobile Network Operators directly. In comparison to CDN pricing, they are all very expensive and built to solve some major experience issues relating specifically to mobile network issues. Lots of value to a mobile network operator in using solution X with the existing infrastructure, as opposed to building out more capacity for the heavy consumption which comes with mobile media.

    Next, all the main CDN providers have different flavors of services but none really have any mobile specific solutions as far as I am aware. Limelight say soon and kudos to them for moving in the right direction. However, to the other CDN:

    ‘Yes Forrest’.
    CDNs do deliver packets, be those streaming packets or http download packets, but I am afraid it is not as simple as ‘Packets is but Packets does’.

    Mobile is ‘like a box of chocolates’.
    Lots of varieties, lots of flavors and lots of stuff that some just don’t like. So when you look at it as just delivering a packet over Ip with no intelligent mobile solution then, ‘ Like a box of chocolates you never know what you gonna get’.

    Now your CDN has delivered a packet to another network that is useless. Some data which your content provider had to pay to deliver and the mobile customer had to pay to receive. Forget the poor experience – people have to pay. Maybe, the network rejects it, often the mobile device cannot support it or the conditions are not suitable and the packet is a wasted resource. Why care if you you get paid anyways.

    I do not mean to criticize but you need to ‘Run forrest..Run’ because mobile is coming at the CDN market and there is a need for mobile specific services on a CDN.

    The long list of vendors trying to sell to CDNs proves this point. It simply annoys me when people think they know mobile. They think it is simple and the same as the standardized desktop platform.

    There are many people who say they can deliver to mobile but do not understand the technicalities and complexities of this platform. This is the main barrier to mobile service adoption and the reason why mobile data services have been a disappointment. It simply does not work on my network, my device or at my particular location at this particular point in time even though it worked here yesterday. Mobile specific solutions solve this issues.

    We therefore need a CDN with mobile solutions. A combination of CDN performance and scale with mobile specific services. Most importantly we need services priced at CDN rates and not mobile solution rates.

    In the current environment with the opening up of mobile networks expect to see more ‘Off deck’ content. Adoption of these open services can be accelerated by a Mobile CDN solving these two main issues:

    i. Experience issue – device support issues, network blocking issues, connectivity and bandwidth issues.

    2. Price issue – adoption is heavily influenced by pricing.

    A Mobile CDN can bring around some serious benefits to mobile services adoption. Removing many barriers that exist today in a similar way to what happened with delivering video to the desktop environment with lower pricing, huge scale and improvements in performance.

    Finally, experience matters on mobile the same way as it matters on the PC. If a Mobile CDN can bring about consumer focused solutions with a better richer experience then all of us will enjoy the benefits. Even Forrest.

  8. I see your column attracts a lot of vendors judging by the attacks on the CTO of Bitgravity. The critics here seem to think Barrett Lyon is scoffing at the viability of video for the mobile market. I think he was scoffing at the idea that you needed new products or services to do it when CDNs are *already* delivering video and music to mobile devices.

    *I* criticize Ms. Higgenbotham’s doubts of the viability of this market. There may be 5,000 different mobile platforms out there today. But you can guarantee that not one of those devices will be in production 3 years or even 18 months from now. The driver for change will be video of course. One commenter correctly points out the huge success of the iPhone.

    In this case the capability to stream video cost efficiently to mobile devices will drive the development of those devices. No need for technology vendors to figure out configurations on 5,000 platforms. The platforms will conform to what the market demands.


  9. To my defense, I did tell Stacey that there were proprietary CDN’s that power networks such as VZ’s vcast, etc. However, my viewpoint is that all these devices will be (if not already) connected via IP on the general Internet, thus there would be no need for a proprietary CDN.

    This is the case already with a lot of these devices, thus a “mobile CDN” really becomes custom content for a mobile device. It becomes special encoded versions of a video, audio file, ring tone, etc.

    The only major custom design work a CDN should be working on for a “mobile CDN” would be designing security interfaces to content so ring-tone purchases can be secured. But guess what… That also already exists in the general Internet CDN world.

    So, with some background I stand by saying a “mobile CDN” is nothing more than marketing fluff if a CDN such as LimeLight networks and not a phone company is building it.

  10. “none of them have gotten much traction inside the U.S. so far.”

    I’ll just speak on behalf of Volantis at this point, since I’m a Volantis employee. Volantis’ software is currently used by all four of the major US telcos and is used extensively in live production by three of them. I believe the AT&T and Sprint logos are on our website.

    Note that like several of the other vendors you mention we also sell to non-carrier customers, and we have a number of those in the US.

    “Some have made limited progress with carries [sic] in Europe”

    I’m not sure what limited means, but more than 35 carriers, worldwide (in Asia, Africa and Australasia as well as Europe and America), are using Volantis for either their rendering platform or their entire mobile portal, and in multiple territories where those carriers have international presence.

    For the rest of the list, I admit there are several I’ve never heard of, and several I’ve heard of but never encounter, but I think that Mobixell (a partner of ours), Vantrix, Ortiva (with whom we’ve also worked) and Vidiator have numerous customers worldwide.

    Happy to discuss this directly, should you so wish. I’m assuming from your blog and from the list that you’re most familiar with the streaming technology/rich media transcoding side, to which we’re adjacent rather than central (hence the partnerships mentioned above).

  11. There are lots of mobile CDN solutions out in the market, but to date none of them have gotten much traction inside the U.S. so far. Some have made limited progress with carries in Europe and the list of those selling to the CDNs and carriers continues to grow and includes:

    – Ortiva Wireless
    – Vantrix
    – ByteMobile
    – Volantis
    – Vidiator
    – Uvumobile
    – Transpera
    – ROK Mobile
    – Thin Multimedia
    – RubberDuck Media Lab
    – Mobixell

  12. Stacey Higginbotham

    @Shai, it will be interesting to see as this nuanced model evolves, how much carriers will end up charging for their part in all of this. I worry that their current 30-percent-to-40-percent slice is pretty hard on the developing ecosystem.

    @henricks and johng, Barrett Lyon at BitGravity is no technical slouch. I submit his personal web site for your consideration:

  13. Interesting editorial and subsequent blog comments:

    The rub is – everyone is correct – transcoding – the ‘secret sauce’ of CDN is in fact rescource hungry for the uninitiated – manpower, software, hardware, servers etc.

    However there is medication available to ease the pain – one of which is a web based video transcoding service – soon to come out of the closet at Streaming Media East this month – VIXOLA – a fast and inexpensive service for transcoding from and to any format including H.264, Flash and 3GP.

    In their alpha testing the service has proven to be extremely easy and efficient with scaleable resources for any medium to large CDN player.

    Mohsen Nasiri COO is culled out of the fiber optic industry and presents an extremely attractive alterative to heavy internal investment for current or future CDN’s and Social networks – [email protected]

  14. I’ve worked for a CDN, a transcoder and a tier 1 wireless carrier. CDNs fundamentally deliver bits faster, cheaper, better. The mobile user experience issue is much bigger than just faster delivery. Sure CDNs can deploy nodes inside wireless carrier’s networks. This is no different from their current deployment strategies. BTW, content is still choked by gateways and OTA bandwidth.

    Transcoding is a whole other animal and has been a fictitous panacea for over a decade. Before considering pushing transcoding to the CDN edge, let’s see it actually work at a single origin host.

  15. I have to say the CTO from Bitgravity is wrong. This is already a huge market but the problem is price. Lots of specialists in mobile but no real scale for any solutions. Bitgravity may deliver some mobile content but where does it come from? Probably an expensive mobile solution provider.

    CDNs have traditionally solved this cost and scale issue. Obviously bitgravity are not going to able to do this with a CTO who does no understand where his mobile content is produced, transcodes or rendered.

    Huge market for Dillithium, Limelight and those who have some knowledge in this space.

  16. I see this is as part of an ongoing trend: Carriers partnering with software vendors to add features or services to their offerings. Comcast & Zimbra, Teliasonera & Soonr, Rogers & Spinvox.

    The old “smart pipe vs dumb pipe” debate is maturing into something more nuanced. Carriers are realizing their value is A) Maintaining a stable delivering mechanism for voice & other content, B) Assembling the right package of features and content, with the right pricing for their demographic, C) Managing a billing and tech support relationship. Given this, it makes sense to outsource the actual features/services/content to 3rd parties as much as possible.

  17. hendricks

    The CTO of bitgravity obviously has no idea what would be requred to deliver a good experience to the mobile audience. The reason why the mobile web has not really taken off is due to poor user exeriences. This is why the iphone is so successful as it provides a good experience similar to other platform. This is proof that mobile is gonna be huge. Users expect this type of media experience even on mobile. Simply put mobile is complex with little or no standards with huge costs for delivery. A CDN can bring allot to the market. Sounds like Limelight have it right and readying themselves for the wave.

    Who would agree that mobile will not going to become the most common platform for web based consumption? Mobile is already the most common method for accessing the web in many Asian countries. Also if you consider the fact that many poorer nations have no wired infrastructure then you know mobile will be the dominant web enabler. A CDN has the ability to help bring content to a world population that is only available through mobile. How about helping to changing the world? Scoff if you don’t understand mobile and smile do.

    Limelight – great to see some vision and they obviously have a good take on mobile.
    Bitgravity – need to put Technical back into the C(T)O. I bet you this CTO does not even know how to utilize his mobile to watch any video.