|By Bruce Crutcher, Tim Dodge||
|March 12, 2006 11:15 AM EST||
As consumers we've become an impatient lot. We live in a world where "on-demand" is a part of our everyday lexicon. Whether it's our travel, our meals, or our money, we just want it now. This is the case with our television as well. Enter the hottest new product offering to be marketed by video service providers: Video-on-Demand (VOD.)
Don't underestimate it - VOD is big business for the cable and telecom companies offering it. Use of VOD services is growing at an astounding pace. Cable companies use VOD as a weapon to thwart the advances of competitive satellite providers who lack the two-way communications infrastructures to provide true interactivity. And what a weapon it is - major deals with networks and studios are making compelling high-quality content available to consumers at the touch of a button - whenever and wherever they want it.
As VOD becomes a standard product offering, there are key features that resonate with consumers, features that can make the difference between a successful VOD rollout and a flop.
Critical Success Factors for VOD
Content is king in the world of VOD, and the availability of popular programs and movies drive the use and growth of on-demand services. Adoption at the consumer level is strong - Comcast served up over 1.4 billion VOD sessions to consumers in 2005, and use of VOD services keeps increasing. With ever-increasing content options to choose from, the sky's the limit for VOD growth.
Huge libraries of exciting content are both a blessing and a curse. With all that compelling content out there, navigation can become a chore. A dilemma of sorts has developed - as more content becomes available, the ability to locate the desired content or browse available options becomes more difficult. Service providers must arm consumers with powerful navigation tools that let them find content quickly and efficiently. Otherwise the value of all that great content can be lost in the shuffle.
The Most Important Factor of All
Content and navigation aside, there's another factor that can make the difference between customer satisfaction and customer defection - reliability. The greatest content lineup can be quickly rendered useless if delivered by a VOD platform that's unstable or unreliable. Unreliable content delivery hurts everyone - the studios, the advertisers, the content providers, but most important, the consumer. Early VOD rollouts were rife with high session failure rates and cumbersome navigation tools - both of which frustrated users a lot. Today, consumers expect VOD experiences that rival Internet uptimes in terms quality of service (QoS).
In an effort to improve QoS dramatically, service providers have invested billions of dollars in their infrastructures to give customers the high-quality experiences they expect. The result is network infrastructures that can deliver the so-called triple play - voice, video, and data - with the reliability and predictability consumers are willing to pay for. And customers have come to expect the same quality no matter what the service. In the eyes of the consumer, video services used for entertainment must meet the same exacting standards as the voice and data services used for business or communications applications.
To be deemed acceptable, VOD services must meet the height of QoS standards. Whether pay or free events, consumers who enter the VOD domain do so on their own volition, and demand a higher standard for quality than with broadcast TV. If a consumer "invests" time or money in a VOD event, that event better be delivered as promised. If service is disrupted or quality compromised, the value of the content delivered is negated. Service providers are demanding ever-increasing reliability from the VOD platforms they deploy.
The bar has been raised for every stakeholder in the VOD food chain. And the pressure can be felt most by the providers of the components that do most of the work - the streaming and storage components of the system.
Benefits of Linux
Video on Demand is an extremely demanding application for today's computers and networks. Session setups and teardowns can happen at alarmingly high rates during peak usage. The Linux operating system provides an ideal platform on which to build VOD systems. The robustness and reliability offered by Linux in general - and specifically Linux with high determinism enhancements - provides an outstanding foundation on which to build highly complex applications that can take advantage of the inherent multithreading capabilities of Linux.
Linux being a "leaner" environment means that programs and the operating system often take up less space on hard drives and require less processing power and less RAM. This means more of the system's resources can be devoted to the heavy lifting of VOD processing, providing the best of both worlds - a powerful VOD system with the reliability and cost-effectiveness to enable business success.
Linux also offers the flexibility of allowing features to be omitted from the running kernel that don't relate to the task in hand. A good example in a video server streaming 2400 streams is that the CD-ROM driver can be omitted since accessing the CD-ROM during the streaming process can adversely affect performance and thus customer satisfaction.
Another example of why Linux is such a good fit for VOD is the ease of integrating drivers as loadable modules. This means new and enhanced drivers can be added quickly and easily to meet changing market requirements. This modular approach lets developers rapidly adapt to service providers' constantly evolving needs.
The general availability of high-performance drivers for high-speed networking and high-bandwidth IO controllers facilitates intercommunication between the various VOD server components and systems to insure continued improvements in VOD performance, streams per server, and ongoing quality of service.
It's More Than Just Software
Software benefits are only one part of the story. Another aspect of Linux that's a strong value-add is the General Public License (GPL) that it's licensed under. Software development philosophies tend to diverge over time due to the often-opposing pull of different interests and the experimental tendencies of developers. The divergence of Unix development in the early days is a good example of that.
But forking isn't always bad because it lets different camps try out different ideas independently (and sometimes competitively.) The power of the GPL is that it lets the key advances of successful divergent development pathways "re-join" the mainstream so that the body of software collectively known as Linux can advance at a much faster rate. This "parallel processing" strategy shortens development timelines for many complicated applications. Ultimately it's the end users who benefit from such an approach. The process effectively speeds up software evolution.
As a related side effect, the code's visibility allows for much better independent peer review, and so it tends to be of higher quality than that produced by a team who knows that nobody will ever see their code.
When Will It End?
In 1965, Intel co-founder Gordon Moore observed that the number of transistors per square inch on integrated circuits had doubled every year since the integrated circuit had been invented. This observation evolved into Moore's Law, which says data density will double every 18 or 24 months for the foreseeable future. Moore's Law has held true for all this time and is expected to be valid for at least another two decades.
The effect of Moore's Law trickles down into virtually all computing applications and VOD is no different. There's an expectation in the VOD marketplace that this translates into rapid advances in server density and throughput. In the past five years on-demand servers have improved from dozens of streams per RU (rack unit) with hundreds of hours of content storage to thousands of streams per RU with tens of thousands hours of content storage. As VOD use grows, this trend is expected to continue well into the future.
One of the greatest assets of the Linux operating system that is sometimes overlooked is the large and growing support community and discussion forums that address issues almost as fast as they appear. New problems are resolved in this Open Source environment faster and more efficiently than possible in the traditional proprietary source code world. After all, the Linux community consists of tens of thousands of enthusiastic contributors who are eager to help solve code issues as quickly as they may emerge. It's doubtful any single company has as many development employees available to not only solve problems in existing code but constantly develop innovative new software solutions.
Boasting high demand for throughput, Video-on-Demand requires an operating system capable of supporting a large number of threads or processes efficiently along with large memory configurations. Linux is ideally suited to the task.
Entertainment's recent evolution to the on-demand business model is a driving force in the development of applications featuring increasingly efficient scheduling and larger memory models. The latest kernel versions of Linux deliver the performance needed for serving up huge amounts of widely varying content to the tens of thousands of simultaneous VOD users in a typical deployment.
Linux and Video-on-Demand are changing entertainment to make it more accessible to viewers, reshaping the way people watch television. It will be exciting to see how VOD grows and evolves over the next decade thanks to a powerful Linux core.
|SYS-CON Italy News Desk 03/12/06 12:01:49 PM EST|
As consumers we've become an impatient lot. We live in a world where 'on-demand' is a part of our everyday lexicon. Whether it's our travel, our meals, or our money, we just want it now. This is the case with our television as well. Enter the hottest new product offering to be marketed by video service providers: Video-on-Demand (VOD.)
- WebRTC Summit Silicon Valley Call for Papers Now Open
- DevOps Summit Silicon Valley Call for Papers Now Open
- Fanning the Flames of Agile
- Select the Right Partner for Privileged Identity Management
- WSO2 Guest Speakers at WSO2Con Europe 2014 Will Examine Technology Developments and Best Practices Enabling the Connected Business
- Call for Papers for Cloud Expo 2014 Silicon Valley Opens
- An API Strategy Is a Business Strategy
- Docker + Stackato: The Perfect Workload Portability Solution
- Cloud Infrastructure for the Real World
- DevOps Summit Power Panel | Is DevOps Really Changing How IT Is Working?
- Big Data Expo 2014 Silicon Valley Call for Papers Now Open
- SaaS Represents the Commoditization of Business Function
- Global Financial Firms Can Effectively Address Technology Risk Guidelines
- WebRTC Business Models: Building a Web-Based Telecom Co
- Enterprise Cloud Analytics and Business Intelligence
- More Mainstream Businesses Depend on Open Source
- AMAG, HP, ImageWare Systems, March Networks and StrikeForce Discuss Security Solutions in SecuritySolutionsWatch.com Interviews
- MapDB: The Agile Java Database
- WebRTC Summit Silicon Valley Call for Papers Now Open
- DevOps Summit Silicon Valley Call for Papers Now Open
- Strengthen Your Business Reputation with Every Application Interaction
- Intelligent Systems in Transportation
- PEER 1 Hosting to Exhibit at Cloud Expo New York
- Learn How You Can Easily Extend Your Infrastructure with Microsoft Azure
- The i-Technology Right Stuff
- After Ubuntu, Windows Looks Increasingly Bad, Increasingly Archaic, Increasingly Unfriendly
- Linux.SYS-CON.com Exclusive: Linus Discloses *Real* Fathers of Linux
- SCO CEO Posts Open Letter to the Open Source Community
- A Closer Look at Damn Small Linux
- Linus' Top Ten SCO Barbs
- Netscape Co-Founder's 12 Reasons for Growth of Open Source
- Where Are RIA Technologies Headed in 2008?
- Linux.SYS-CON.com Exclusive: What Would UserLinux Look Like?
- *POINT - COUNTERPOINT SPECIAL* What's Wrong with the Open Source Community?
- Why Recovering a Deleted Ext3 File Is Difficult . . .
- Introducing "Cooperative Linux" - Linux for Windows, No Less