DISASTER RECOVERY through archive management

In 2005, “disaster recovery” became a household term in North America. On the heels of catastrophic events, including tsunamis and earthquakes in other parts of the world, the tremendous damage inflicted on the Gulf Coast of the United States by Hurricane Katrina and other storms demonstrated both the significant amount of destruction caused by natural disasters and the overwhelming rebuilding process that follows.

In many industries, disaster recovery and business continuance have been topics of discussion for some time. These recent, wide reaching disastrous events have once again brought this necessary planning to the forefront. The problem is complex and can be costly. In the auto industry, for example, IT departments take pride in ensuring daily backups are maintained for parts databases, e-mail servers and network shared drives. This ensures that in the event of a database server failure, car production lines will be minimally impacted. The important thing to understand in this example is that the real value of the auto makers' business is not these databases or e-mail servers, but rather the actual work product — the automobiles they produce.

If auto makers had to store every automobile they manufactured in a single warehouse, their value would be extremely vulnerable to potential natural or other disasters. Although this “super” automobile storage facility is not a realistic option, it highlights what millions of broadcasters, news agencies and content owners do to store the lifeblood of their business — their videotape and film assets. It is difficult to comprehend how many of these assets have been lost in the aforementioned natural disasters.

Disaster recovery and business continuance differ fairly significantly in their end goals, but they share a common progression path. Disaster recovery is just that — the ability to recover in the event of a disaster at a facility. Business continuance is the ability to remain on the air as a disaster is occurring, even if in a somewhat diminished capacity.

Content protection in a traditional non-digital environment is a demanding task, requiring duplication of every piece of videotape or film in a broadcaster's possession and subsequent transfer and storage of that material at a second site. By moving a known “good copy” of a media asset to an alternate facility or storage site, a company can prevent a calamity such as a fire from destroying years of valuable footage. This approach clearly falls into the category of disaster recovery.

While this solution is effective, it is hardly efficient. The mere process of copying material is labor-intensive, particularly if proper quality-checking processes are put in place to ensure the best possible quality. In this scenario, every hour of broadcast material might require several hours of work to replicate. For most broadcasters, this amounts to years or even decades of duplication work. Even if a company were to make this investment, the matter of sorting, finding or repurposing specific material from within the tens of thousands of hours on tape or film presents a whole new challenge. In essence, this proposition is a no-go.

Replication of content in the analog world may not be feasible, but the broadcasters' transition to digital, file-based operations has facilitated much easier, much faster and much less-expensive means of content duplication. Digital files can be replicated at faster-than-real-time speeds, with little or no human intervention. However, even this manner of protecting content takes an initial investment of time and money.

Digitization of broadcast assets

The adoption of digital encoding technology and the migration of content off legacy analog storage media (videotape, film, etc.) is the fundamental prerequisite to any effective and realistic asset-protection scheme.

The primary focus in the transition of assets to digital format is ensuring that the best possible quality digital file is created during the ingest process. This is the most labor-intensive step and is best performed only once. Intelligent and reasonable analysis must play a part in the balance between digital asset file size (and, therefore, storage capacity requirements) and the digital encode quality selected for these assets as there are digitized. The quality and condition of the source material or videotape format itself can factor into the determination of encode parameters, as it is not necessary to choose a uniform encode rate for every asset. The secondary goal is to start this transition to digital as soon as possible. If the process is not started immediately, the problem continues to get worse as more and more analog assets are moved to storage shelves in the basement archive.

The natural degradation of aging video and film media is an internal disaster in the making, rendering many thousands of hours of content contained in global archives unrecoverable each year — many times unbeknownst to the asset owners. With the adoption of digital storage technology, the shelf life of stored media is no longer a concern, as advanced archive management solutions will facilitate the automatic migration to newer storage formats on demand. This migration process is a pixel-by-pixel replication of the asset ensuring an exact duplicate is made, significantly faster than real time with little or no manual intervention. This migration process for media lifespan protection can also include digital asset reformatting (transcoding and/or rewrapping) to conform to newer, emerging encoding standards, thereby guaranteeing transparent compatibility with newer editing, newsroom and playout systems.

Once these best possible quality digital files have been created, archive management systems can automatically replicate the content, providing various levels of disaster recovery or business continuance depending on infrastructure investments — with no further human intervention required.

In addition to the obvious benefits of content protection and easy access for repurposing are the significant density benefits of digital storage technologies. For example, encoding your assets at 50Mb/s (typically accepted as archival quality but likely overkill for older analog sources) would allow nearly 20 hours of digital content to be stored on a single LTO3 data tape with a street price of approximately $100, all in less than half the size of a large format Betacam cassette. So, by leveraging your archive management system, high-quality duplicate copies of nearly 20 hours of your valuable content can be stored in a compact space for less than €200 — plus the cost for shelf space at your remote facility, of course.

Still, this challenge of digitizing thousands of hours of legacy analog content can be a daunting task, making the decision to commence the transition a difficult one. In the end, the broadcaster or media group must make decisions about the importance of content and prioritize their digitization process accordingly. The digitization of all a company's media assets represents an enormous capital investment, which could take several years to accomplish. Carving out a percentage of those archive hours worth the time and expense is really the only viable option. Start with the most important, significant, valuable or relevant content identified on the archive shelves, and work backwards from there. Once this decision is made, likely with the help of a facility librarian or some kind of content list or database, key footage such as important political events can be preserved for future access before other less significant material.

With these critical media assets digitized and protected, the broadcaster can begin to digitize material on demand. While footage of the pope or a past president may have fallen into the priority category, digitization of other material — such as tsunami footage, made more newsworthy by recent events — can be driven by the course of current events and the business itself. Rather than the librarian or archivist releasing these assets to producers on videotape as they are required, simply prioritize their digitization and make them available from your archive management system as digital assets.

Protecting your digitized assets

Steps toward asset protection can be taken one by one. Each step offers incrementally increasing protection, starting with simple disaster recovery and eventually leading toward true business continuance. The common element to the systems is the primary ingest, media asset management and archive. (See Figure 1.)

Obviously, there is an incremental cost associated with the move from step to step:

  • Automatically replicate assets (on digital storage media) at the main facility, and store these duplicate copies in a secure and geographically diverse location. (See Figure 2, Option A.)
  • Replicate a digital storage infrastructure (i.e. data tape library) at a geographically diverse location, and have the system electronically replicate assets via WAN connectivity. (See Figure 2, Option B.)
  • Reproduce the main facility archive management infrastructure at a geographically diverse location, and have the system electronically replicate assets via WAN connectivity. (See Figure 2, Option C.)
  • Reproduce the main facility broadcast infrastructure, staff the geographically diverse location, and mirror the playout operations of the main broadcast facility. (See Figure 2, Option D.)

It is important to note each of these steps can be adopted in phases, over several years, leveraging the digitization of assets as the first and most necessary step and spreading capital investments over several fiscal periods. For example, the media automatically replicated and stored at the remote facility in the step-one approach can be simply loaded into a second data tape library installed in step two. Step two can lead to step three by the addition of a second archive management solution at the remote site, leveraging the library and the duplicate content created in steps one and two.

Similar significant barriers exist to protecting the auto makers' assets, and they exist as well for the broadcast content-owners. Chief among them is cost. For auto makers to ensure full asset protection, they would have to make duplicates of every automobile and store each in these geographically diverse super garages. It would be far more efficient if the auto industry did not actually have to build a duplicate vehicle for each produced, but was able to protect this fundamental value with little more investment than the additional parking spaces required. That simple and cost-effective asset protection is the value proposition advanced archive management systems offer to content owners who have made the transition to digital.

These systems can be configured to replicate digital content automatically following any one of the schemes outlined above — from simply duplicating the assets on digital media (data tapes, Blu-ray laser discs, etc.) and allowing the owner to store these in a secure off-site location, to having two live facilities actively mirroring each other for true business continuance goals. The important factor is that each of these protection steps can be taken incrementally to minimize the initial investment while still providing key protection for the valuable assets with little more investment than the necessary “parking spaces” for the replicated media.

Some broadcasters cannot afford to house media remotely simply because they require instant access to backup copies in the event of a primary copy failure. A second (or even third) backup copy could be kept on site, but this option adds to both the financial burden and storage space required for content protection.

Major broadcast networks are taking this concept one step further, linking complete facilities cross country, so that if one facility were to go down (along with its program transmission), then the facility on the opposite coast could use its duplicate library to maintain the broadcast output. This is the end goal of business continuance, achieved through incremental and manageable steps. (See Figure 3 on page 42.)

Accessing your digital assets

An archive management solution serves as middleware, working in the background to connect the facility's journalists and other “media consumers” with these extensive digital archives. The archive system often is used in conjunction with a media asset management (MAM) application to allow easy access to the stored digital assets. The MAM system is the forward-facing client application, providing creative staff, archivists, media owners, etc. with advanced search, media identification, metadata association and other advanced features such as “proxy” browsing. Advanced archive management systems also include integrated audio/video transcoding functionality, which can generate frame-accurate, low bit-rate proxy versions of all of the digitized content to facilitate easy desktop browsing of the stored high-resolution content. These proxy versions of the high bit-rate assets are typically passed to the MAM application, which further enhances the user experience when attempting to locate content in their digital archives.

Consider again the example of raw tsunami news footage. Without a digital archive management solution, the journalist or editor must contact the videotape librarian, who in turn will attempt to physically locate and screen all candidate footage (which may span a significant number of tapes) for relevance using his or her subjective judgment. The librarian will then make duplicate copies of the most relevant material to protect the original source media and deliver these duplicate tapes to the requesting editor. The editor will then screen the content once again for relevancy in the context of the story he or she is preparing, digitize the best portions of the footage into the nonlinear editor and produce the finished news story. The amount of work required prior to the actual creation of the piece is time-consuming and inefficient.

In the case where a digital archive management system has been adopted with some level of MAM functionality, efficiencies are obvious and significant. Rather than having to contact the librarian, the editor would simply open a Web browser on his or her desktop computer, access the MAM system and query the asset database for keywords such as “tsunami,” “earthquake” and “ocean wave.” The MAM system would return a candidate list of matching assets stored in the archive management system.

Using the proxy content generated automatically by the archive management system, the editor can directly screen all of the relevant assets for applicable footage, potentially identifying key shots that may have been missed by the librarian in the previous example. The editor then can directly mark segments of the assets using the frame accurate proxies, generating an edit decision list (EDL) of relevant portions of the archive content.

Once all relevant shots have been identified, the editor can simply use the MAM Web interface to request the retrieval of only the relevant segments of the original assets identified in the EDL, leveraging the ability of advanced archive management solutions to perform timecode based partial restore operations. The relevant portions of the digital content identified by the editor will be retrieved from the data archive library, automatically transcoded or reformatted as necessary and delivered directly into the nonlinear editing system. The labor efficiencies gained and vast creativity benefits of a digital archive management system are obvious.

A variety of effective archive management options are available to broadcasters both large and small, and they can offer substantial benefits not only in the event of failure or natural catastrophe, but also in day-to-day operations.

Choosing a system

Any number of systems, running the gamut in terms of cost and complexity, can be implemented to safeguard media assets and realize digital workflow efficiencies. The resources of a broadcast station, television network or other media company clearly will be one of the deciding dynamics. The key is to identify and clearly prioritize the goals of the implementation and not attempt to address every hope and dream immediately since cost and complexity will likely make this initial step impractical or even impossible.

Start small, and partner with an archive management provider who will work with you to incrementally achieve your goals. The first step could simply be to stop accumulating more videotapes in the facilities tape archives with no immediacy to the goal of implementing a MAM solution. The next step could be automatic replication of these digitized assets for disaster protection, and so on. Take manageable and cost-effective steps toward accomplishing your goals.

As more and more broadcasters and media organizations move their assets into the digital realm, the task of protecting that content becomes easier. From that point on, each company must prioritize its assets and the future revenues, real or potential, that content represents. Thoughtful implementation of archive management solutions can provide the means to protect content against loss whatever the source and offer numerous other operating benefits that add efficiency to everyday workflows.

Brian Campanotti is chief technology officer for Front Porch Digital.