We deploy all of our projects in Microsoft Azure. We are a startup company. We have been since the beginning deploying our code in Azure cloud. We are actively using the full capabilities of two cloud services.
The application services are the most valuable in Microsoft Azure. I'm not using them directly but I am using the function and the web applications. I don't need to pay a lot for the maintenance. I do not need to have a DevOps employee.
I am familiar with Google, and everything I was doing was in Google. I had to control my back proxies and do my own configuration files. With Microsoft Azure, it is all easy.
The documentation can be outdated and is not as clear in Microsoft Azure as it is in AWS or Google.
I have been using Microsoft Azure for approximately four years.
The solution is stable.
It's easy to monitor, there's no problem with the code in their development. We had some stability issues, it was not the fault of Microsoft Azure, but it was the mistakes of the developers. It was easy to monitor it in Azure.
Microsoft Azure is highly scalable.
On only one web application I created with one service plan I can connect 10 projects. Imagine one machine can run 10 projects at the same time because whatever the project has more requests or demands, it will scale its machines and auto-scale down. This is why it makes it affordable. I can control which project is demanding more computing power or storage power.
We have approximately 50 people using this solution in my organization. It is mainly back and front developers.
We are drastically increasing our usage because last year we were running 13 projects, and now 19 projects. In 2022 we are trying to double our team.
The support could improve. For example, Python is needed for Microsoft Azure, and the lack of documentation for the community is a problem. If you are a Python developer Microsoft Azure released an update to Python at the end of May 2021. Theoretically, I can use it with Python, but if I have a problem I need to call a Microsoft engineer to solve it. It takes some time. However, I did receive very good support from the Microsoft engineers to make my system production-ready, but language support for Python and other languages is coming late.
I have used other solutions, such as Google and AWS.
When comparing these solutions to Microsoft Azure, AWS has better documentation, Google's cognitive services, and predictions give better results, and Microsoft Azure has the best UI. If you want to reach the database of Google, there is no such SQL manager UI. Microsoft Azure UI is easy to use and has great tools.
The deployment's easy and the pipelines are easy from the Azure DevOps, everything is integrated and it has good security. The overall setup is extremely easy.
We have a team in London that is maintaining the solution. We have 450 backend developers at this site and approximately five people are looking after all the maintenance and admin roles. We have database and developers administrators that are giving access to the people for the production maintenance.
I am very happy with the solution.
Thanks to the training I have received from Microsoft Azure, which cost £60,000, I'm up-skilling all my team for the certification, databases, and machine learning tool. Every month I'm receiving approximately £1,000 from the training credit for the up-skilling.
It is simple to start with Microsoft Azure if you know the application life cycle. You can try so many things without any cost because of the serverless system. You will not be charged for any request at the beginning. For example, you are creating a function application in Azure the first 10,000 requests are free. It is great because you can anyone a developer to test anything.
They're not using very heavy machine learning systems, the system is generally cheap. For example, they are giving a free month trial and a developer can't finish it personally if they do not use a large computing machine.
At the moment they are adding new features faster than I expected. For example, they have Python support but five years ago there wasn't any Python support. They were slow at the beginning but now it updates very quickly. For example, the community services for the low code, no code power platforms, and the power platforms.
I spoke to my developers, machine learning engineer, data engineer, and data scientists and told them please use the auto ML or the community services better. As a London business user site team, we can create all the virtual agents and AI builders with the no code platform for the machine learning models for the power applications. The auto ML is very powerful and you don't need to be an expert in machine learning.
I rate Microsoft Azure a nine out of ten.
