Pipeline Publishing, Volume 4, Issue 7

	This Month's Issue:
	On The Horizon

Over-The-Top Services

article page | 1 | 2 | 3 | 4 | 5 | 6 |

plants will be pressed to keep up with bandwidth capacity challenges as OTT video explodes; yet they are competitively accelerated by have existing relationships with media companies and an understanding of this market. Telcos with fiber and switched VDSL probably can support capacity demands, but may founder in understanding how to leverage legitimate media content. We are not used to seeing cabelcos and telcos collaborating, or even waiting for each other to catch up; instead they compete.

Lastly, every time the operators train up staff & engineers and then downsize, non-traditional operators (including OTT service vendors) get human resources: experienced human capital that can build networks and services not under the economic control of traditional Operators. Google is not buying an Operator because they bought the talent instead and can now build a network tailored for their needs.

SIP vs. IMS

SIP has recently been called the “third great protocol of the Internet.” SIP enables a wide variety of services to be composed and delivered in near-real-time. 3GPP and IMS standards groups choose to embrace SIP instead of something like SS7. SIP now fuels IMS, but does SIP need IMS? Indeed, no. SIP is itself architecturally neutral and IMS is a specific (and rather elaborate) architecture. SIP can be deployed in P2P mesh models, in heavy-edge-client models, and in platform embedded widget models. Today, SIP is used somewhere in all of these. Also, development environments like Astrix are providing a bridge from software code to telco services – with development times in days rather than months. This is showing dramatically that users can get fancy services and receive converged data/media/voice without service providers. But will these be quality, managed services?

Some say this dual trend means the death of IMS, but IMS and 3GPP also include a Policy QoS model. This may serve to be the important feature which gets IMS deployed after all. And this is very closely related to the management of quality networks. Using identification of service type, a known registry of user needs and desires, and a flexible policy model of specifically how to tune the network to deliver consistent quality services is, in our view, the most solid business model for today’s service provider. This should allow the provider to bill for consistent QoS for specific services delivered to specific users with specific product tariffs. Who the service provider bills then does not matter - as long as someone will pay… and if no one pays,

Is the service provider opting into the same “bad guy” image of the digital rights distribution companies? Or will the users just break DRM again wasting everyone’s development efforts?

then the Wild West, best-effort environment ensues. And the very teenagers driving OTT services may drive the best-effort network into effective gridlock. Then media rights owners and advertising driven portals, and even household parents, will step up to pay for the value of predictable quality.

So we see the advent of OTT services not as a threat to service providers, but potentially as an ironic opportunity. Success, however, will go only to the agile service providers who plan and act effectively, and not to those who continue to stick with their traditional tendency toward bloated network and management projects.

Key OSS/BSS needs

So, since we are not likely to see our industry end in ruin, let us return the implication of OTT services on OSS/BSS.

Current peak-usage metric models for managing and engineering network traffic probably do not reflect the assumptions and use patterns of current OTT services. Longer, more sustained-rate transmissions may become the norm instead of the exception. The old statistical models used to compute reasonable over-provisioning of access vs. core capacity will not hold in this new OTT service world. Billing based on inexpensive access capacity depends on over-provisioning for profit. But profits will be squeezed as access broadband becomes a near constant up and down stream – a data fire hose of un-billable services, so over-provisioning will disappear and access charges will increase – or become service dependent.

Current firewall and security services are not designed to support the many widespread connections (to other user machines) that are set up by P2P mesh applications. These mesh networks are used by such dominate gaming products as World of Warcraft (WOW) to download updates while playing continues. These distributed updates generate equal upstream and downstream traffic. If WOW users will buy high-end machines to play competitively, someone will pay for tuned, consistent quality network connections to the WOW servers.

Information paths routed across many separate carriers with guaranteed service will be common place. There will need to be an increasing collaboration of information among these networks. Security gateways will

article page | 1 | 2 | 3 | 4 | 5 | 6 |

© 2006, All information contained herein is the sole property of Pipeline Publishing, LLC. Pipeline Publishing LLC reserves all rights and privileges regarding
the use of this information. Any unauthorized use, such as copying, modifying, or reprinting, will be prosecuted under the fullest extent under the governing law.