Smart spaces, whether its buildings, campuses, transportation hubs, or even cities are no longer just the talk of technologists. They have become real, but they are also highly fragmented, and it is difficult to deploy applications to them. Izuma Networks is changing this.
The rapid merge of IoT devices, powerful AI edge silicone, and greatly improved wireless connectivity is making it possible for smart spaces to finally come to fruition. However, the more sophisticated these environments get, the more important it is to have a modern application deployment strategy at the edge.
Deploying applications at the site allows for highly durable, reliable experiences which can be specific to the site location.
Location-specific applications allow for:
- Logic which understands architectural or configurations specifics
- Local control of the application while at the site. For instance, if the site experiences an Internet outage or power failure
- Occupant experiences that make sense for the specific venue
Previously deploying and managing unique applications at each site might have been a near-impossible task. Continuous deployment strategies, such as Izuma Cloud, which uses Kubernetes APIs, make the management of these types of deployments very feasible.
Izuma Networks Edge-as-a-Service is a continuous deployment model. This model means software remains up to date and that adjusting software is a natural and frequent part of operations.
If you are developing smart devices for smart spaces, look into our solutions for IoT and gateway devices.
Izuma Cloud combined with Izuma Connect or [Izuma Edge can significantly decrease your development timeline, helping you rapidly get products to market.
Vertical or point solution vendors will often pitch both a hardware device and a cloud product. Such solutions will serve their users well for a specific use case, but having many of these solutions in a building will result in many siloed services, which are difficult to use to make a specific actionable decision.
Tying these siloed solutions together means developing another service, running on a cloud provider, which in turn talks to each siloed service’s own cloud APIs. Then when a decision is made to take action at a site, that cloud reaches down to control some device in a building.
This approach (left) has a lot of problems:
- It’s very latent. All legs of the journey require communication across the Internet. Each one will require a couple of roundtrips to complete its own API call - of around 10-40ms. So time could be close to a second just to get the data needed to make a decision.
- It’s dependent upon a lot of service providers: each cloud for each device and all the providers they may use.
- It’s dependent upon the Internet working in the space itself.
On the right is a pure edge approach. Here the application making the decision is running in the building. It talks directly, over LAN or other means (such as RS-485) to each device. The entire process is local. The application is deployed, monitored and upgraded by Izuma Cloud. While the Internet is needed to deploy the application, the app can run even if the Internet is disconnected in the building.
The most likely approach integrators will need to take is a hybrid one. There are going to be certain devices that designers want to use, which have their own dedicated cloud, and there is nothing that can be done about this. At the same time, other devices’ data can still be processed locally.
This approach will gain designers:
- A locally running application, which can still make decisions with disconnected Internet.
- A lower latency decision-making process. Many devices can still be talked to locally
- Devices that have their own dedicated cloud are still part of the system logic
The devices which have a dedicated cloud service will be processed by a pod, usually running in the cloud part of Izuma Cloud. This pod will prepare the data and then make it available to the edge application.