Over the years, one of the problems I hear a lot from broadcasters, especially for news and sports production workflows, is: “I have operations in different locations. I’d love to have them work as one unit and be able to collaborate closely with one another, but at the same time, in case a location gets disconnected from the others, I still want it to be able to operate totally normally.”
It’s a rather traditional question of balance between centralized and distributed systems. The more centralized it is, the easier the collaboration will be, but the more dependent it is on proper connectivity. The more decentralized it is, the more robust it will be, however the sharing and collaboration capabilities will be more limited. There is no “one-size fits all” solution to this question. The optimal choice will vary from one system to another, depending on a number of parameters such as:
- The reliability and bandwidth available for the network connections
- The criticality of the operations in case of network disconnection
- The collaboration requirements between the different locations
These are not easy compromises to make for the technical teams designing and maintaining the system, especially since they are not easy to accept from the editorial teams who would want to “just have it work seamlessly.” And today’s users, who are used to having access to all their tools from anywhere, are becoming even more demanding.
A new toolset
Luckily, there are a few developments in the recent years that can help better manage these situations. The first one is related to the pure IT evolution. Networks become faster and more reliable for lower costs all the time; virtualization technologies facilitate the availability of distributed resources. In effect, some of the technical constraints have become less crucial than they once were.
The second aspect is related to the ability to make smart use of MAM software layers in each of these systems in order to optimize the workflows and user experience. Most modern MAM-based solutions (such as Dalet Galaxy) offer some level of import/export and intersite exchanges. This can be at the metadata level and/or the media level. This provides a powerful toolbox that allows users to build more flexible intersite workflows. Using these capabilities, some of our customers, for example, have decided to handle the problem with rather creative ideas, making use of the power of these intersite exchange tools under the control of the workflow engine.
One example is a group of stations that decided to organize their operations in the following way:
- They created “regional hubs” that act as production and playout centers. Each of these hubs runs a local system that can work in full isolation.
- Some smaller “satellite locations” are geographically dispersed. Each one is associated with the most relevant “regional hub,” and works on the same system.
- They have also built a “central collaboration system” where the content from each of the regional hubs is sent via “smart replication.” This regionally created content is automatically copied over to the central collaboration site based on workflow rules (this collaboration system is actually receiving rich metadata updates and proxy media from each of the regional systems).
This “central collaboration system” is running on a virtualized infrastructure. It allows all users across all sites to collaborate and share content, based on rich metadata and low resolution streaming video. But when they actually need to exchange the HD video material, all heavy lifting is handled with direct file transfers using network acceleration methods.
Others have chosen different architectures, but the key to making all this work is the flexibility of the exchange tools and the ability to have them “smartly driven” by the workflow engine.
Dealing with multisite workflows is not as complex and painful as it used to be. Making use of modern MAM layers (and the toolbox they bring with workflow management driving “intelligent” intersite exchanges) can drastically improve the workflows while allowing us to avoid some of the difficult trade-offs we used to have to make between centralized and distributed architectures.
If you are facing such issues, I would recommend to start looking at this with a fresh mind. As I said earlier, there is no “one-size fits all” solution, but with the right tools becoming increasingly available in today’s world, finding a flexible solution is well within our grasp. Of course, we’d be more than happy to help you find the solution that fits best for you.