structure of a distributed video wall
Decentralized video wall management systems are network based systems, which are composed of stand alone units (input, output nodes), in contrast to classic video wall controllers (so called splitt PCs, or matrix controllers).
All sinks and sources communicates via standard Ethernet network protocols and the data is transported via common network switch technology. There is no proprietary hardware needed (like expensive matrix switches) and the network can be easily changed or expanded.
On the output side (sink) every display is connected to it own stand alone hardware (decoder - meetwise DC1) to display all available sources. Input wise (source) hard- and software sources are support to get the signals of various sources into the IP network.
The whole systems is managed by one or multiple master units.
Since all nodes (units) are stand alone - even if one of them fails, the whole system still runs with only small outtakes like one missing screen or source. So the resilience against hardware or power failures is much higher than on conventional video wall systems. Redundancy can be built in on multiple switches, multiple master and redundant network connections.
Last but not least, additional nodes can be added with ease, so the integration of further screens or sources is possible every time, without limitations at very reasonable costs. The system will grow with your needs and your application.
This example show a setup consisting of a 3x2 video wall and two single displays. The following sources are used:
- one multi screen workplace PC, its 3 DVI signals are encoded with 3 encoder modules
- a IP TV streaming Server
- 4 analog cameras, encoded by a network encoder (e.g. AXIS or BOSCH)
- a transcoder to connect older IP based cameras, which are not able to produce multicast streams
- a master unit and a user station
- a tablet for additioanl control, which is connected via WiFi
24/7 usage / resilience / redundancy
Since every node of the system is working stand alone, a failure of one node only compromises this single point of the system. So in worst case, one output or one source is not usable anymore. This offers a very high resilience against failures compared to classic matrix switches.
There are multiple, well known and tested ways to design an Ethernet network in a failure tolerant way, e.g. by using multiple, redundant switches, cables, NIC teaming and so on. All these tools and strategies can be used to build a network to the degree of safety you need.
Alle nodes of the distributed DisplayDriver are equipped with their own soft and hardware watchdogs to constantly monitor their own health. Due to the usage of a very "slim", trimmed down Linux OS quick boot times and exceptional reliability are made possible.
The decoders are having their complete OS and software on an external USB dongle. In case of a hardware or power failure, the technicians just need to replace the hardware unit and plug the USB dongle into the new hardware.
The master units can operate in n-times redundancy. For example, if you are using 3 master nodes in your setup, even if 2 of them suffer from a hardware failure, the last one still can fully operate the complete system.
With all the measures to raise the availability and reliability of the system, the transparency of these tasks is important for the service technician. For normal users, these jobs are not visibile, they just keep on using the system even in worst situations.
Meetwise MiniDisplayDriver Downloads
- meetwise distributed DisplayDriver overview (english, PDF) (132.6 KiB)
- meetwise distributed DisplayDriver Datasheet Master M1 (english, PDF) (282.1 KiB)
- meetwise distributed DisplayDriver Datasheet Decoder DC1 (english, PDF) (279.8 KiB)
- meetwise distributed DisplayDriver data sheet Decoder DC1.2 (english, PDF) (415.7 KiB)