Cloud Providers Comparison for IoT Applications: Amazon vs. Microsoft vs. Google

Welcome to our new blogpost about the comparison of cloud providers! After a long examination of these three cloud providers and their services, their features, advantages and disadvantages in the matter of IoT applications were worth to be shared. If you haven't read the further blogposts yet, we highly recommended you to read them first.

Microsoft Azure Part 1

Microsoft Azure Part 2

Google Cloud Platform

Amazon Web Services

Image 1: Cloud Providers

There are several articles about this comparison between these three providers, but nearly all of them are focusing on big data, data storage, cloud functions and so on. Our key points comprise on IoT applications, device-cloud connections, device-cloud messaging, IoT security, sensor data storage and real-time data visualisation by using cloud functions.

Engineering is a trade-off. They all have specific advantages and it is better to compare them according to use cases. They all use MQTT and HTTP protocols. AWS is better on gathering data when you want to use large databases, Google has advantage on data usage and analytics and Azure is better on computing and data visualisation. Pricing is also different on these three providers. So, better to search the best provider for your special needs instead searching the best in all cases. Don't forget that they don't have huge differences.

IoT Cloud Services

Some cloud providers look like they have more services then the others, but some of them correspond to more services. In the end, irrespective of the number of services, their functions are highly comparable. So, don't let the names make you confused and let's investigate these services now.

Hub

Tool lets the "things" connected easily and securely, simplifies the device-cloud and device-device communications.

Analytics

Data visualisation, data storage, log data monitoring, device-cloud messaging and analytics can be done with following services.

Security

The security of cloud-device connections can be satisfied by different methods on these providers.

Based upon these comparisons and former blogpost experiences, let's complete the pros-cons list:

AWS:

Advantages:

  • Dominance of the cloud market with the biggest market share for more than 10 years.

  • The array of the functions and cloud services are not only already wide, but also still growing.

  • Ready for use "Enterprise-ready" features.

  • Rich documentation about the services and use cases.

  • Using FreeRTOS makes everything faster and easier.

Disadvantages:

  • Complex usage of services and functions.

  • Overwhelming amount of services.

Azure:

Advantages:

  • Better documentation for the end user and practical applications.

  • Advantage of gathering projects and users that already using Office, SQL Server, SharePoint, .Net etc.

  • The faster connection between cloud and the device and easier alliance.

  • Powerful on putting the apps on cloud.

Disadvantages:

  • Implementation of real-time data visualisation is complex.

  • Incompatibility with Safari web browser.

  • Complex user interface.

  • Starting a new service has longer creation time then other providers.

  • Some connection problems between PowerBI and Azure.

Google:

Advantages:

  • Strong theoretical documentation of the modules.

  • Compatible with all the browsers.

  • User friendly dashboard.

  • Connection stability between Stackdriver and IoT Core.

  • Practically it is good to have Mongoose OS or Zerynth OS when you are dealing with examples.

Disadvantages:

  • Weak practical documentation for the end user because of missing examples.

  • Lack of IoT services in contrast to AWS or Azure.

  • It is a little bit harder when you don't have Mongoose OS or Zerynth OS.

  • More complex device authentication leads user to generate the public and private keys singly.

Conclusion

That's pretty much it! In most cases, you can never go wrong with choosing any of these providers. Amazon has the highest market share, Azure has lot's of tools, Google has stability... For IoT applications, we can say that Google has a little bit more way to go in order to being competitive against Amazon and Microsoft. Nevertheless, it is always important to find "the best" for your special needs according to your project.

As ThingForward we are focusing on every field on IoT and stay tuned for more blogposts!

Eren Mert Irez