Media Server Technology: Karl Paulsen

Tempering the Cost of the Digital Transition

Broadcast group owners continue to explore alternatives to operational practices in order to offset the costs required to make the transition to digital. Alternatives already include such ideas as centralized or jointly operated master controls, whereby the amount of hardware and labor necessary for operations is significantly reduced by sharing resources at one location among multiple remote facilities. The technology needed to efficiently operate in this new environment will undoubtedly change the direction of products and systems manufactured and implemented during the next several years.

The possibilities for interaction over great geographic distances continue to grow as the electronic separation between the facilities begins to shrink. Keeping that concept in perspective, it is important to consider what impact today's decisions relating to choices in servers, formats and storage architectures have on the future of media content storage and delivery.

When looking at the computer data industry, digital content storage continues to grow at a phenomenal rate. In just the past few years, this growth rate has increased at an unbelievable rate of approximately 60 percent per year.

CONTINUING TREND

We fully expect that as the broadcast industry continues its shift toward video file server technologies, the degree of data storage and the types of hardware available will continue to grow as well. Still, we caution that there are obstacles and in many cases outright boundaries to the successful implementation of a totally dedicated digital television architecture. In focusing on the technology associated with media-based file servers, we find there are two areas of significance that will require serious attention in the coming relatively short time frames.

Fluid connectivity, which will be necessity need to include cross compatibility of digital formats, is one of the key hardships that must be overcome before successful joint or shared operations will become totally successful. The other area, which we'll call storage migration, encompasses the long-term preservation of (and compatibility among) the myriad digital data developed once the analog methods of videotape and associated media become useless as legacy recording and reproduction hardware is retired.

Today the available connectivity solutions are riddled with a multitude of choices and in turn a variety of known and unknown associated costs. Now that the digital domain is firmly here, there are a variety of stable and accepted methods of dealing with the local, intrafacility connectivity solutions.

The distribution internally of digital media can certainly be controlled on a local basis. Yet once a digital connection breaches the boundaries of the immediate local facility and into the area beyond the metropolitan area, the equation changes significantly.

COST DELEGATION

Some organizations have justified the initial and additional ancillary costs, such as servers and landline (including fiber) leases, by delegating these costs with accounting principles aimed at reducing direct labor costs and by the amortization of hardware in new methods. However, as the boundaries expand, the still unknown and often variable parameters used to justify these choices are many and may adversely affect certain decisions, should market conditions change or technology shift direction.

When choosing a means of connectivity associated issues and choices need to be identified. In the case of point-to-point delivery methods, such as the use of fiber optics as a transport medium, it is insufficient to state that fiber capability is already in place without knowing, in depth, how that fiber can or may be used.

If the world were interconnected by masses of dark fiber strands that anyone could utilize for their own specific purpose, like the point-to-point microwave radio model broadcasters understand well, then a fiber connectivity solution would be a slam-dunk. But the real world is not so gentle. Most fiber-optic links, where intended for video uses, must consider such elements as the input format and hardware, the compression and modulation methods, as well as the switching that occurs when passing through multiple central offices, drop points or other points of presence - including cable headends.

RECEIVE AND TRANSMIT

Getting the broadcast signal from the master control terminal control area to the fiber pickup point may utilize a completely different scenario on the receive end vs. the transmit end. While recognizing that high-quality video transport in larger metropolitan areas (such as Hollywood) is in place, on the whole there are only a modest few choices for wide area, long-distance distribution of isochronous broadcast-quality video with the fluidity we've grown to know from satellites or microwave.

We are, however, on the cusp of some very interesting changes in video delivery from point A to point B, and beyond. Now that the Internet pipe is widening, as the cable companies install broader fiber-optic paths and even as some utilities have seen the light by providing open rights of way for video connections, things will begin change more rapidly.

CAPITALIZING ON COMPRESSION

Even today it is possible to capitalize on the combinations and benefits of video compression in conjunction with digital media server technologies for the store and forward of high-quality, short and even longer form video content over nonconventional methods of delivery. The ability to e-mail, over an IP address-based Internet connection, a 12 Mbps, MPEG-2 4:2:2 Profile Main Level product from a videotape to a server, or even the desktop for approval purposes, is here today.

It will not be long before you'll be able to justify the costs associated with the statement, "If you need it faster, just install a bigger pipe." The new delivery models being developed today may change how advertising and commercials will be made and delivered and in turn how the operations of tomorrow will be modified.

Besides contribution video, broadcasters are now seriously considering across-the-state and other regional distribution of their program stream for cable headend or translator repeating. Alongside this is their desire to drop in local ads and even local programming without building or staffing satellite station operations.

THE REALM OF POSSIBILITY

This concept is not beyond the realm of possibility. The technology now exists that will permit the preloading of a remote video server, in nonreal time, with unique local content and interstitials, while on the same fiber-optic strand, delivering the stations' programming in real time across the region.

Combining such concepts will permit local spot availabilities to be sold and aired without additional staff or the enormous amount of hardware that the typical remote station located in a smaller market within the same DMA might require today. With digital servers and automation, the quality will be better and the reliability will be higher.

Yet in order for some of these ideas to become reality, we will need to refine solutions to the previously mentioned problems of cross-compatibility and long-term stable data storage formats. It will not be possible to utilize the capabilities of nonreal-time delivery of data to servers if the formats, packetization and compression methods change at various gateway points along a video or media pathway.

This remains one of the advantages of satellite-based video. We will not refute that broadcast satellite delivery is stable, understood and (other than occasionally messed up by operators) remains a reliable and dependable means of contribution and transmission. Yet these operators are jumping on the multipurpose bandwagon as well, and we expect ancillary services to blossom as the thirst for content delivery expands.

With the variety of alternative and new solutions available, including ATM, Internet, Internet 2 and dedicated private compressed services, the most significant tasks that should be undertaken in this new dawn of digital delivery is the systemization and quantification of these capabilities. The universal standardization of signals and paths for the interchange of moving media will depend upon how the manufacturers of the receiving and transmission equipment, including storage-based server systems, continue to develop their product lines.

It is exciting to witness the changes coming across the industry as certain individuals take the risks to develop the hardware and software infrastructures necessary for these concepts to succeed. This year will see a great challenge to reduce costs and improve capabilities in the digital domain of television broadcast.

Karl Paulsen is the manager of systems integration at Synergistic Technologies Inc. He is the author of the book "Video and Media Servers: Technology and Applications" (published by Focal Press) - a compilation of the past several years of this column and its applications to the broadcast industry. Contact him via e-mail at: kpaulsen@STIDigital.com