IoT Cloud Platforms Solutions: An Overview

Companies that leverage IoT systems are looking for appropriate solutions to their needs. They are looking for analytical suites to analyze the data and make sense of it. They are looking for solutions to manage the devices. In short, they are looking for a platform that gives them all the functionalities needed to run an IoT based system from a single pane of glass. This is where IoT Cloud Platforms come into play.

This article surveys the top IoT Cloud Platform solutions, those offering a PaaS (Platform as a Service) for the IoT ecosystem. We will discuss which services the platforms offer, their costs, and evaluations. Finally, we’ll touch on the factors for choosing an appropriate IoT platform. In a follow-up blog, we’ll deep-dive into performance testing techniques for these platforms.

What are IoT Cloud Platforms?

IoT cloud platforms are comprehensive solutions that provide all the necessary functionality and services for an IoT implementation.

IoT Cloud Platforms vs. IoT Development Tools

Before we get started, let’s clarify some definitions. In addition to “IoT Cloud Platforms”, you may have come across the term “IoT Development Tools”. What’s the difference between Platforms and Tools?

This could get tricky, because sometimes you will see solutions such as Arduino, which is used for programming IoT devices, alongside Oracle IoT, which has DB capabilities. Obviously, these are two disparate products, and comparing them is like comparing apples to oranges. Furthermore, there is a whole range of products in between them, such as Eclipse IoT, which is an open-source framework encompassing several tools that can be used to create an IoT platform. 

So how do you draw the line between Tools and Platforms? While Tools target specific tasks, such as the development of an edge device, Platforms offer a comprehensive solution for management, analysis, data gathering, and storage. Some companies that have a relatively straightforward IoT implementation and use common protocols may opt to use a platform as their solution. Other companies, specifically in the Industrial IoT area, may need a customized solution to perform a specific task and be integrated with their legacy system. In this case, an open-source tool such as Eclipse IoT may serve better.

This article focuses on IoT platforms because they reside in the internet cloud environment. As such, testing the performance of these systems and their ability to handle the load generated by a multitude of devices is extremely important. How to actually test these platforms will be the topic of a follow-up article. But here we will gain an understanding of what features and services that IoT platforms provide.

IoT Cloud Platform Overview

Providers

Interestingly, and quite surprisingly, too, there are hundreds of IoT cloud platforms to choose from. It seems like many major players in IT and mobile are offering some sort of IoT service. This goes to show how IoT has become mainstream and ubiquitous. For this article, I will look at some of the top players, but obviously, the list of IoT Platforms is much longer. Here are the big players in this field:

• Particle
• IBM Watson IoT platform
• Salesforce IoT Cloud
• Cisco IoT Cloud Connect
• Oracle IoT Platform
• AWS IoT Core
• Azure IoT Hub
• Google Cloud IoT

We’ll review them later on. But first, let’s understand what IoT platforms provide.

Features

IoT cloud platforms are usually characterized by providing these types of services:

• Device management
• Connectivity
• Security measures
• Data storage
• Data analysis
• Message broker
• AI capabilities
• Rules engine

Let’s explore what each one of these services entails: 

Device Management

Device management is the most common feature of each platform provider. The need to centrally manage a multitude of IoT edge devices is a must for any IoT system. This includes tracking the devices, updating the firmware, securing the devices, authentication, and functional monitoring.

Connectivity

Connectivity deals with how the edge devices connect to the cloud provider. Some devices connect using a SIM card and others by Wi-Fi, Bluetooth, or radio. When dealing with SIM cards, the platform needs to manage activating the SIM, changing data plans, monitoring the service, and deactivating a subscription. This is called “connection life-cycle management”.

Security Measures 

The security of IoT systems deals with making sure that the data transferred from the edge device to the server is not compromised. Security systems look out for “rogue” devices that have been hacked. If data is transmitted by Wi-Fi, then the connection must be secured. These systems are monitored for cyber-attacks, DDOS, and authentication infringements. 

Data Storage & Data Analysis

IoT devices generate a huge amount of data, and this must be stored in databases for future analysis. This data could be sensor data, environmental data, positioning data and more. It is important to also capture the context of the data, its origin, temporal information, etc. for useful analysis. Because of the large quantity of data needed to be analyzed, commonly called Big Data, many providers store the data in NoSQL databases such as MongoDB or HBase. To ensure performance at scale, it’s essential to validate these systems—learn more in How to Load Test HBase with JMeter.

Message Broker

All IoT platforms have an integrated message broker for receiving transmitted data. Some of these brokers are proprietary and others commercial. The two most common protocols for data transfer are MQTT and HTTP. If the IoT implementation is using a different protocol, then it is important to make sure that the platform supports it.

AI

Making sense of all the data being gathered by IoT devices requires Artificial Intelligence (AI). Most platforms have AI capabilities as well as Machine Learning (ML). ML is useful in surveillance systems and Industrial applications. 

Rules Engine

Finally, Rule Engines are a way of utilizing the information gathered by IoT solutions and using them to take actions. These actions are the business rules that govern the system implementation. Not all IoT platforms have an integrated Rule based engine, and sometimes the appropriate solution depends on the particular use case.

Now that we’ve understood the features of IoT platforms, let’s dive into a description of each vendor.

Vendors

Particle

Some platforms, such as Particle.io, offer device development tools and even their own devices, alongside data management and storage capabilities. It is a real end-to-end solution provider. 

This is an example of a company that started off providing development tools for edge devices. Then they provided cloud services to connect the devices, device management capabilities, and finally, analysis tools. It was a bottom-up approach.

But Particle.io stands out in this category of IoT platforms. Most platforms don’t deal with the edge device layer. What is also unique about Particle.io is their extensive support for all types of cellular connectivity. This is because they provide fleet management solutions based on their proprietary EtherSIM card.

Salesforce IoT Cloud

The Salesforce IoT platform, created in 2015, is based on Open-Source Apache tools: Kafka for message handling, Storm for Big Data handling, Spark for data processing, and Cassandra for data storage. For business analytics, they use Einstein Analytics, which merges sensor data with ERP records. The Salesforce solution focuses on the customer and augments the IoT data with its CRM capabilities. 

As an example, one case study of ferry tracking, where the ferry is the IoT device, emphasizes how data collected from the ferries can be used in conjunction with “people” data. So, if the ferry is delayed, passenger contacts can be notified by SMS. Or knowing a passenger’s location and ETA can be used for an up sale. The impression is that the Salesforce platform is geared towards solutions where people/customers and the enhancement of their experience is the main objective of the data collection. 

Cisco IoT Connect

Cisco is a leader in computer networking, and it is no surprise that they are leveraging this in the field of IoT and specifically Industrial IoT. The Cisco IoT platform handles the secure connection of IoT devices to the network, manages devices using the cellular network for connectivity,  tracks devices and their functionality, and provides additional device management services.

You could say that Cisco is more of an IoT infrastructure solution provider. They offer hardware for edge devices, edge computers, routers, and gateways. From the software perspective, they have tools for monitoring and managing the network of devices and hardware making up the system.

The Big 5: Amazon, Azure, IBM, Google, and Oracle

All the top five IT companies offer the most comprehensive IoT platforms on the market. They have a full range of features in their suite of products. Additionally, they have support for a relatively new technology called “Digital Twin”.

A Digital Twin is a virtualization of a real-life product. It could be an airplane, wind turbine, washing machine, car, or any physical product. With this technology, sensors on the real-life product transmit data to the cloud. Using this data, a simulation can be generated to study the product and analyze its performance. 

Pricing

Pricing is an important factor in choosing an IoT platform vendor, but it is usually not the most important. Surveys have shown that the most important factor is functionality. However, price is also a factor when comparing two or more comparable products. Unfortunately, comparing the cost of each product is difficult because of the very different price models that are being used for each one. 

Here are some of the cost factors used in the pricing models:

• data volume
• number of connected devices
• computational events
• device messages transmitted
• message size

Here are some examples of how prices are determined, as taken from the vendors' internet sites. Keep in mind that these plans change from time to time, so this is only an indication.

In general, there is no sure way to know how much the monthly cost of using an IoT platform will be. Only after running the system for a couple of months, you will be able to get an approximate cost. So, the best you can do is estimate.

Criteria for Choosing an IoT Cloud Platform

In an article published by the IEEE Internet of Things Journal, the authors listed 21 factors to consider when choosing an IoT platform:

The authors recommend an approach where the most important factors are determined and then each vendor is ranked by their strengths in those areas. 

Bottom Line

IoT Cloud platforms are used to implement IoT systems and provide common functionality and services. All the solutions offer device management, connectivity support, a message broker, data storage, and analytics. Each vendor has a unique focus or leaning typically based on their main field of expertise. Cisco is more appropriate for Industrial solutions, Salesforce for integrating with CRM, IBM for small businesses, and so forth.

Customers can usually choose from a suite of capabilities according to their specific needs. The different pricing models of each vendor makes it difficult to compare costs, however price is usually a secondary factor in choosing a vendor. Finally, choosing an appropriate IoT platform is a complex task that needs to consider multiple factors and special considerations of the underlying implementation.

BlazeMeter is dedicated to helping customers implement the appropriate testing methodology for their IoT solutions. BlazeMeter offers a cloud-based SaaS application used for functional testing, performance testing, and API monitoring. Many plugins are available by BlazeMeter to integrate with other systems such as CI/CD pipelines and APM monitoring systems. IoT customers from various sectors are using BlazeMeter for their testing needs.

START TESTING NOW

Related Resources: 

•  IoT Predictions for 2022
• IoT Continuous Testing with Jenkins and BlazeMeter
• Testing in the Cloud: Benefits, Best Practices, & Examples