Large Scale Playout

Whatever we call it ('linear streaming'?) TV is here to stay. It is then appropriate to ask ourselves of the most efficient way in which we can achieve this with the available technology.

As with the transition from tape to disk years ago, IP distribution is reshaping the TV industry and particularly the way playout is managed. Companies like Vector 3 have evolved their products throughout the changes in the TV industry over more than three decades and are well positioned to forecast the changes in the landscape of playout that will come about due to the new and improved technologies (WAN connectivity and the cloud).

Analysing the change from tape to file, we can see that this provided the environment in which production and playout could be geographically distant from each other without any real-time penalty in the delivery process. In the beginning the opportunity was not realised as it was assumed that video files were too big to be moved over WAN. Increased availability of bandwidth and the development of techniques to move large files have made the transport of content worldwide a relative straightforward process. This has favoured the concentration of a large number of channels in large scale playout centres, reducing unit costs by increasing scale.

The TV industry is now ready for the next change that will separate the operators from the playout. This change will maintain the advantages of large playout facilities whilst allowing an, as yet unknown, degree of flexibility. The devices that originate the signal can be grouped and located as required and the operators no longer need to be co-located. One particular form of creating the signal – video servers in the cloud – is already requiring remote operation but the concept can be extended to all other of ways of originating the signal. One operator can be monitoring channels that are in many different locations as if they were in adjoining rack rooms. We must get used to this as we got used to files being downloaded from remote locations.

Vector 3 is one of the companies at the vanguard of these new approaches thanks to its broad experience in all the revolutions and distractions of the past 30 years. Having worked "from day one" with non-proprietary hardware, LANs and WANs are the natural place for its software to work. For many years, its MediaManager software has been feeding storage, firstly from cart machines, then from remote studios and now from virtual locations. FileIngestor automatically performs a range of operations related with the ingest. In the past, ingest only needed to cross the gallery, then came the LAN and now content can be in one of the many physical locations or virtual disks wherever content is stored. MediaCacher deals with the last step, years ago, less than 30 meters, today potentially thousands of kilometres. Using this suite of software tools, it is possible to build a playout facility that spans the whole planet, with the origination of channels being both in specific remote locations or in the cloud.

The trend towards large playout facilities has led to the development of techniques that allow operators to manage and monitor more and more channels. These techniques are very close to what is needed for the next step. Workflows have been standardized and automated to maximize the productivity of operators. In this new context, the operator is no more the driver of the processes but the manager of issues as they arise. The playout system must present only the relevant data and focus their attention on necessary interventions. An important point is that the operator needs to act in advance to avoid problems becoming an issue. The remote placement of the playout has increased the need for anticipation in solving incidents that may require communication across time zones. To streamline the process and to make the information relevant, customisable time horizons can be implemented, so as to focus the operator's attention and avoid extraneous warnings.

As important as it is to present timely warnings, it is only the first step in developing the process. Everybody involved in playout – or in any other complex automated process – knows, "monitoring" not only includes recognising the warnings but a continuous proactive search and effective problem solving. The dashboard must guide the operator to the root of the incident and help him to identify the origin. Once the cause is fully known, "monitoring" includes solving or mitigating its effect as much as possible. Vector 3 MultiChannelMonitor has evolved to perform these three functions that are encompassed in the word "monitoring." From its dashboard, the operator is able to drill into the data coming from all the available sources and react according to the situation.

The ability to separate the control and monitoring from the physical or virtual place where the emission is carried out can lead to another approach in certain cases. It would be possible to decentralize this control and monitoring so the company hosting the physical or virtual devices used to broadcast, can allow its customers act as operators from their own locations. All these combinations can be changed dynamically depending on the time day or the day of the week.

Once the TV industry has been able to grasp the full possibilities of this revolution in playout fulfilment it will be ready for this leap forward. Some things have changed forever and the rack rooms will be no more a static presence one door away, but a set of remote datacentres, out of sight. Other things will remain as will the need for skilled operators and specialized playout software.

www.vector3.tv

This article is also available to read at BFV online here, page 34, as part of Playout & Distribution feature.

(JP/LM)