Moving Media Storage Technologies

Applications & Workflows for Video and Media Server Platforms

Karl Paulsen

• Publisher: Routledge
• Year: 2012
• ISBN:
  9781136023057
• Edition: 1
• Topic: Lingue e linguistica
• Subtopic: Studi sulla comunicazione

Complex media storage computer systems are employed by broadcasters, digital cinemas, digital signage, and other business and entertainment venues to capture, store and retrieve moving media content on systems that will preserve the original integrity of the content over time and technological transition. This book provides detailed information related to the concepts, applications, implementation and interfaces of video file servers, intelligent storage systems, media asset management services, content distribution networks, and mission critical platforms.

A tutorial and case example approach is taken to facilitate a thorough understanding of the technologies, using numerous illustrations, tables and examples. The text and appendices are designed to provide easy to access valuable reference and historical information.

.A focus on the media serving concepts and principles employed at the enterprise level

.Practical and technological summaries of the applications and linkages between media asset management and storage technologies for studio, television, and media production workflows

.Illustrations, standards, tables, and practical summaries serve as handy reference tools

1

INFORMATION, DATA, AND VIDEO

The process of digitally storing moving images and sound begins by converting once analog video and audio signals into a stream of discrete representations of binary numbers corresponding to sampled levels of chroma and luminance (for video) or for audio, correlating the changes in dynamic air pressure, which have already been converted into electrical signals as they were captured by transducers over time. In the digital media world, this string of binary information is represented as data and is often referred to as bits and bytes.

The recording function collects, organizes, and retains this data in a format that can be recovered for reproduction as visual information on a display or audible information on a loudspeaker-like transducer. In between the collection and the reproduction phases, the data is retained on a media form. The retention platform is referred to as storage, which can be both a process (i.e., the process of actually capturing or placing the data onto a recordable media) and the physical media itself (e.g., a magnetic hard disk, a lineal videotape, or an optical disc such as a CD-ROM or a Blu-ray Disc).

KEY CHAPTER POINTS

• When does data become information
• Qualifying structured versus unstructured data
• Requirements for the storage of moving images
• Identifying analog composite and component video signal sets
• Digital video recording explained in terms of its components
• Elements of digital video (bit depth, colorimetry, and formats)

Data Structure

In general, data may be classified as either unstructured or structured. The factor for classification is determined by how the data is stored and how it is managed. Rigidly defined data is much easier to access and manage than random sets or elements that are arranged in nondiscriminate structures.

Rigidly structured data is traditionally relegated to databases that are built and maintained by a fully fledged database management system (DBMS). Unstructured data is considered “everything else”; however, there is an area in between, which is neither rigidly structured nor completely unstructured. This gray area is called semistructured because it has a reasonable amount of order or grouping but is clearly not consistent across an enterprise, a department, or even an individual.

In business, more than 80% of an enterprise’s data that is typically used for non-media-related purposes is unstructured. This raises huge concerns for the enterprise as unstructured or semistructured data consumes an enormous amount of storage and is extremely difficult to manage.

Taking an ad hoc approach to data structures for media-related data extends the complexities well beyond what the traditional business space has had to face—with the issues and concerns growing even larger. From this, the business and technologies of storage have become a science.

Information

Data that is created either by individuals or by enterprises needs to be stored so that it can be accessed and easily recovered for processing at a later time. There is no reasonable purpose for data unless it is capable of being presented in a meaningful form. In the business world, the knowledge and intelligence derived from this data are called information.

Before the current digital age, data were collected on printed paper and stored in filing cabinets or placed on library shelves as books or periodicals. Information could be gathered from this format because there was a structure for accessing the data that could be configured in a repeatable and definable classification. The Dewey Decimal Classification (DDC), also known as the Dewey Decimal System, was developed by Melvil Dewey in 1876, and is an example of a proprietary system that is used for library classification. The DDC is one methodology for arranging data in a structured classification.

Today, information is categorized by myriad methods each tailored to a product, an application, a location, or a service. Much of the information presented is dependent on a form of indexing. The indexing uses the concepts of tagging to put the data into a structure that is usable by others. Search engines have become the foremost means of discovering that information.

Given the enormous and growing amounts of media-related information, the value of intelligent storage and the systems associated with searching and retrieval are being elevated to new proportions. Managing the storage systems, and complimentary distribution paths such as networks and transmission devices, is becoming a significant part of what was once an area dedicated to the information technology (IT) sectors only.

For the media and entertainment industries and the public and private business sectors, these concepts are rapidly changing, resulting in a new paradigm for those systems associated with the storage of media-centric data and information.

Storing the Moving Image

Film was and remains one of the most notable forms of storing visual information either as a single static format (as in a photograph) or a series of discrete images on a continuous linear medium (as in a motion picture film).

For at least the past three decades, moving images and sound recordings have been stored on various media including wires, plastics and vinyl materials, rigid metallics, or other polymers, but primarily these recordings have been captured on the most well-known magnetic media, simply referred to as “tape.”

The information contained on this lineal magnetic recording media is often segregated by the type of recorded information contained on the media, that is, sound on audiotape, visual images with sound on videotape, and for the computer industry bits representing binary information on data tape. Over the course of time, the format of these audio-, video- and datarecording processes has moved steadily away from the analog recordings domain toward the digital domain. This migration from analog to digital recording was made possible by a multitude of technological innovations that when combined into varying structures, some standardized and others not, formulate the means by which most of the audio, visual, and data are currently retained (see Fig 1.1).

However, these are all changing, as will be seen and identified in the upcoming portions of this book. Videotape as a recording and playback medium is gradually, but finally, moving away from the traditional formats of the past nearly half century.

Figure 1.1 A selected history of recording and physical media development.

Digital Video Recording

Digital video recording is a generalization in terminology that has come to mean anything that is not expressly an analog recording. These much belabored references have worked their way into every facet of the industry, compounded even more by the widespread introduction of high definition and the transition from analog over-the-air transmission to digital terrestrial television (DTT) broadcasting.

Clarifying 601 Digital Video

It goes without saying that digital video brought new dimensions to the media and entertainment industry. For the video segment of the industry, this began in earnest with the development of standardized video encoding in the 1980s whereby interchange and formalized products could store, in high quality, the video characterized by its original description as CCIR 601, which later became known as ITU-R BT.601 when the CCIR, International Radio Consultative Committee (French name: Comité consultatif international pour la radio), merged with others, and in 1992 became the ITU-R, International Telecommunication Union- Radiocommunications Sector.

As with many emerging technologies, technical jargon is generated that is filled with marketing hype that makes the concept “sound good” regardless of the appropriateness or accuracy of the terminology. Digital video recording was no different. With its roots embedded in analog (NTSC and PAL) structures, digital video terminology has become easy to misinterpret. In the hope that the readers of this book will be given the opportunity to “get it right,” we will use the terminology that is standards based, or at least as close to the industry accepted syntax as possible. To that end, we’ll start by describing the migration from analog component video through to compressed digital video storage.

Analog Component Video

Prior to digital video-encoding technologies being used for the capture and storage of moving media, videotape transport manufacturers recognized that the image quality of moving video could be better maintained if it was carried in its component video structure as opposed to the single-channel composite video format used throughout studios and transmission systems prior to digital (e.g., as composite NTSC or PAL video). Not unlike the early days of color television signal distribution in the studio, where video would be transported as discrete components throughout portions of the signal chain, the process of video interchange between analog tape recorders was pioneered by the early use of “dubbing” connector interfaces between component analog transports.

Analog Component Signals

In the era of early color television broadcasting, the impact of the legacy monochrome (black and white) signal sets remained as facilities moved from hybrid monochrome and color systems to an all color analog television plant. Those component video signals that comprised the technical elements of video signal systems consisted of both composite signals and discrete sets of signals including the primary red-green-blue (RGB) triplet set of color signals. Depending on the makeup of the video processing chains, additional horizontal and vertical synchronization (H and V) signals, composite sync (S), and color subcarrier (3.58 MHz color burst) might have been distributed throughout the facility.

Often, these signals were carried on individual coaxial cables throughout the television plant. As color television systems developed further, the number of individual discrete signals diminished, except for specialty functions such as chroma keys. The concept of component video signal technologies reemerged in the early 1980s, enabled by the desire to carry other sets of individual channels of component analog video on either pairs or triplets of coaxial cabling.

Two fundamental signal groups make up the so-called component analog video signal set. The signal set described by the three primary colors red-green-blue, better known as simply RGB, may be considered the purest form of the full bandwidth set of video signals. In a three-channel color television imaging system, a set of signals is generated from the light focused in the camera’s lens and is passed to the prismatic optic block (see Fig 1.2). When the field image passes through the prism and is optically divided into three images, each corresponds to one of the elements of the RGB color set. Depending on the electrical makeup of the image, which may be a CCD or a CMOS imager, the electrical signals generated by the imager are amplified and their components are applied to electrical systems in which they are eventually scaled and combined according to the signal standard to form a television video signal.

Analog component video signal sets found in video systems may include more than just the primary RGB triplet of signals. These additional component signal sets, designated as RGBS, RGBHV, or RG&SB signals, are often used when connecting computer-generated video graphic signals to displays configured for video graphics array (VGA) signals and beyond.

In its “pure” format, RGB signals should be considered fullbandwidth representations of optic-to-electrical energy. In this format, full-bandwidth RGB signals are rarely used in video processing or video production. RGB signals generally do not use compression and in turn do not impose real limits on color depth or resolution. As a result, they require a huge

Figure 1.2 Component video to serial digital video bit stream signal flow, filter, and multiplex (MUX).

amount of bandwidth to carry the signal and contain significant amounts of redundant data since each color channel will typically include the same black-and-white image associated with the percentages of color information for the RGB channel they represent. Recording this type of RGB signal onto analog lineal videotape would be impractical, though not necessarily unimplemented. While the digitization of the “pure RGB signals” would offer the cleanest preservation or integrity of the image, it is fraught with a myriad of issues far beyond the topics of this section.

Typically, when component signals are used in analog or digital video recording, the RGB components are not used but instead a different scaling and combination referred to as color difference signals or sometimes just component video signals are used. Component video signals are scaled differently than the straight pure RGB signals, given that video displays must account for both the human visual system and the physical properties of the display system. Because of these issues, we alter the scaling of the RGB signals to produce a nonlinear set of video component signals that are used in displays, digitized videos, and compression systems. When these nonlinear signal sets are derived, the...