“Designing resilience into the power and cooling systems”

Steve Burgess, technical director at KIT digital Broadcast Systems Integration offers his perspective on designing network operating centres (NOCs).

“No matter how much redundancy is built into a particular site, there may still be occasions when working from the site could be impossible… provision has to be made for disaster recovery”

Operation 24/7 is critical to the success of the business, so care has to be taken to design resilience into the power and cooling systems. An uninterruptable power supply is essential (preferably with redundancy) along with a reliable back-up diesel generator. All systems should be carefully maintained and tested on a regular basis.

The equipment passes programme-critical signals, so redundancy is needed. This can range from redundant power supplies, through to complete spare programme chains in a separate chassis and kept in a hot standby mode.

No matter how much redundancy is built into a particular site, there may still be occasions when working from the site could be impossible. Although it is hoped that these events never happen, provision has to be made for disaster recovery. This leads to the need to be able to control and monitor the traffic passing through the NOC from a remote location.

The operators often have to endure long shifts with some periods of intense activity. It is extremely important to create a suitable environment with ergonomically designed chairs and desks, and comfortable viewing angles to all critical items of equipment. It is well known that uncomfortable operators are likely to make mistakes. Equally, we do not want them falling asleep during a long, quiet night shift.

Historically, control of the various subsystems would be via the manufacturer’s own control system, often switched to the operator’s desk position by a large KVM matrix. For example, there would be a PC or other types of controllers running an application specifically designed to control the encoder multiplex system. This would be supplied by the manufacturer of the encoder multiplex system. The same would apply to the glue system, routing system, antenna system and many other parts of the sent/receive chains. An operator sitting at his desk would not be able to have separate displays/keyboards/mouse for each of these systems, but would use a KVM matrix to switch his single display/keyboard/mouse to the appropriate PC/controller.

This is OK, but the operator needs to be familiar with all the different ways of presenting data and the different control systems as designed by each manufacturer. It is inconvenient however, to teach the operators all the different control systems and many of them will be used very rarely.

System-wide monitoring and control systems (M&C) have been around for some time to control the various systems in a satellite teleport for example, and increasingly the requirement is extended to be able to control more aspects of the NOC operation.

In the ideal world, one user interface would control everything from the bookings, to the hardware switching, to the monitoring and conformance checking, and reporting. Multiviewers have become the norm in recent years and these can be used to good effect to allow increased flexibility in what is monitored and in what form. Increasing sophistication can be built in, alerting the operators to any problems, thus allowing an operator to be in control of many more channels.

Satellite interference seems to be becoming an increasing problem for the owners and leasers of transponder space. More and more sophisticated tools are being developed to help identify that interference is in fact taking place (as opposed to poor operation) and to identify the source. These carrier monitoring systems must also be integrated into the overall control systems if possible.