buchachon - Fotolia


Know when to use cloudless IoT or stick to the cloud

Despite the cloud serving as a touchstone of 21st-century computing, organizations have set up a growing number of IoT implementations without cloud connectivity.

Cloud connectivity has served as an integral part of IoT deployments, but organizations must determine when sending data to the cloud works best and when it would be better to go cloudless.

In many IoT deployments, a variety of machines collect, store and manage disparate data in a large, remote cloud composed of software servers and databases. Cloud computing gives organizations advantages in scalability and efficiency. It can also cut down on development costs when an organization does not need to set up the infrastructure themselves to process, store and analyze IoT data.

However, organizations may not always want to rely on a cloud service for their IoT deployment and would rather use edge or fog computing. Cloudless IoT can bring computing even closer to where data is generated and decrease the latency or improve security.

IoT architects must understand each situation that could call for cloud or cloudless IoT deployments.

IoT in the cloud offers extensive storage

Organizations can connect devices directly to the cloud or use an IoT gateway to connect machines to the cloud. To connect to cloud software, a device first needs to connect to the internet, which can be done with connectivity options such as cellular, Wi-Fi or low-power wide area network (LPWAN). In device-to-cloud connections, transport and application level protocols facilitate the data transfer. The amount of memory and storage a device has on board can also determine if the device must link to the cloud. Some IoT devices have megabytes of storage on board, but many do not.

Cloud services can experience outages and may not offer the reliability that an organization needs.

Organizations that use IoT services might rely on the cloud for a variety of reasons, including:

  • The cloud offers a near-limitless supply of remote data storage, whereas IoT devices store only very limited amounts of data or none at all.
  • Some IoT projects must have extensive processing power to complete tasks, and cloud systems give users access to an array of computing resources. 
  • A cloud system provides users with remote access to control multiple IoT devices and retrieve data from units, sensors and other machine-to-machine hardware. 
  • Bringing IoT software into the cloud makes it easier for users to access and manage IoT applications through an integrated whole. IoT teams can use the cloud software that incorporates the IoT back end to apply security updates and upgrades.
Edge-to-cloud architecture

Cloudless IoT can provide more privacy and security

Not every IoT project requires the cloud. An organization may prefer to run a cloudless IoT project, or even one disconnected from the internet. Cloudless platforms, such as Microsoft Azure IoT Edge or Crosser IoT Edge Analytics, run workloads on edge IoT devices off the grid to increase data privacy and security for IoT implementations.

Organizations might plan for IoT deployments to remain unconnected to a cloud service because:

  • IoT devices might not be able to access the internet to connect to the cloud where they are located, such as in remote facilities or deep underground sites that cannot reliably access cellular or satellite connectivity.
  • Data privacy is a key concern. Organizations can ensure data privacy better when data stays where it's generated. 
  • Employing an IoT system without internet keeps data more secure from hackers and other cyber attackers. Key municipal deployments, such as water or power plant operations, must prioritize data security.
  • Cloud services can experience outages and may not offer the reliability that an organization needs. A cloud system can even disappear if the company hosting ceases to exist or stops hosting that particular cloud software. Avoiding the cloud can ensure the stability of a continuous or very long-term IoT project. Any cloud-based software may become unreachable if the cloud goes down. Numerous outages have happened, including four major cloud software provider outages in 2020.

IoT devices may not need an internet connection to operate, but they do require a link to other machines on a network to automate certain tasks. So, the concept of localized, cloudless IoT presents advantages for certain business needs.

The concept of cloud computing and IoT services evolved commercially over roughly the same period in the early 21st century. Cloud took off when major industry players started offering cloud services.  Amazon introduced its Elastic Cloud Compute service in August 2006. Google followed and officially launched Google Cloud Platform in April 2008. Microsoft Azure launched in February 2010. Today, vendors offer a multitude of public and private cloud systems for use.

IoT technology started to come into public awareness as the "internet of things" in the early 2000s. Kevin Ashton, co-founder of the Auto-ID Center at the Massachusetts Institute of Technology, first used the term in 1999. The introduction of the IPv6 protocol in 2011 allowed the creation of billions more IP addresses, which massively improved machine-to-machine communication. In the latter half of the 2010s, engineers developed IoT devices capable of using NB-IoT and other LPWAN specifications. Alongside Wi-Fi and wired connections, these devices can connect to the internet and therefore the cloud.

Dig Deeper on Enterprise internet of things

Data Center
Data Management