Domorela's Blog: Intranet of Things Private Cloud Service deployment (II)
As said in our previous article speaking about Intranet of Things, after deployment of the TCP/IP next step is deployment of Private Cloud Service that publishes a WWW service that will fit in an environment of a certain size.
In order to deploy a Private Cloud we can opt for the same products offered as SaaS, we can install them in our Intranet with an, usually expensive, On Premise license applicable in large environments. But we also can opt for a product conceived to be used in an Intranet of Things, like Domorela, designed to operate in small, medium and large environments.
Prior to SaaS model, the market offered two types of products to provide solutions to specific environments: dedicated devices and dedicated licensed software. A dedicated device is usually running a dedicated software (firmware) fitted on specific hardware. A dedicated software can be installed on a multipurpose hardware with an Operating System capable to run it.
Specific hardware solutions are cheaper than multipurpose hardware doing same tasks. Think about a PC acting only as a router and compare its price to the one of a router offering same performance. Also, a licensed software requires more time to develop than firmware as usually needs to be compatible with many types of hardware components present on a multipurpose server, so the solution is to be compatible with several Operating Systems (OS), or adopt only one of the existing ones and become prisoner of the OS vendor APIs (Application Programming Interface).
Of course the explained above doesn't take into account Open Source development. If we rely on Open Source APIs things are changing, so its possible develop using Open Source Permissive Licenses allowing us to close our specific code and compile it to be executed on different OS and hardware. If you do that you may contract services and/or donate a part of your profit to the maintainers of the projects you rely on.
Reading the above we can speak about three types of Private Cloud deployments:
- Install a dedicated device, sizing it for our environment, implementing the IoT service we need in our intranet. Suitable in all environments.
- Install an Open Source Software in our multipurpose server that implements the IoT service we need in our Intranet. Suitable only for medium and large environments.
- Install an On Premise license of a SaaS solution in our multipurpose server that implements the IoT service we need in our Intranet. Suitable only for large environments.
Nowadays, dedicated devices are being implemented with Embedded Systems and the combination of those with Open Source APIs is giving a lot of new designs for the IoT, so we can speak about Embedded IoT Systems.
Usually, people are thinking about embedded systems like little low power and low performance devices that are doing a few tasks and become clients of a SaaS installed on a Private Server or Public Cloud. But dedicated devices can be more or less powerful and, given the last generations of microcontrollers, embedded systems too. Hence we can classify embedded IoT systems as:
- End device: any embedded IoT system acting as a sensor for data acquisition and/or performing dedicated tasks such as controlling electronic inputs and outputs. These devices can be IP compatible devices or not, if not the installation must include an IP gateway capable to communicate with them.
- Edge device: any IP compatible embedded IoT system deployed as a gateway to communicate different segments of an IoT network, usually communicates end devices with a centralised service, but can also communicate non-IP segments with IP segments. Should be capable to receive and store data from end devices, send commands to them, and also to communicate with central service for data exchange. May be capable to publish a WWW service.
- Core device: any IP compatible embedded IoT system deployed as a core IoT service. Should be capable to communicate with any other IoT device, storage data and publish a WWW service.
Domorela can be deployed as core or edge device and can be scaled to fit in any installation of small, medium and large size. As in the previous article, below table briefly compares a deployment of Intranet of Things, now from devices perspective, with an x86 computer and IoT software solution with a deployment done with Domorela in small, medium and large environments:
| installation size | solution based on an IoT platform runing on a computer | solution based on Domorela |
| small |
Core device:
Edge devices:
End devices:
|
Domorela device deployed as Core device:
End devices:
|
| medium |
Core device:
Edge devices:
End devices:
|
Domorela device deployed as Core device:
End devices:
|
| large |
Core device:
Edge devices²:
End devices:
|
Domorela console deployed as Core device:
Edge devices²:
End devices:
|
(1) Modbus hardware compatibility already present in Domorela, software module will be included in future version
(2) one edge gateway for each remote IoT site network attached to backbone IoT network