The cloud has changed our software architectures

[ayshakhatun3113]

The cloud brings new services
Cloud providers offer a large number of ready-made services, such as databases , user management, message brokers and much more. This makes the operating environment of our cloud software very different from the environment found in the classic data center.

On-Premises Data Center
When we build software in the Java environment that will be run in the customer's data center (on-premises), the environment is typically an application server or Docker environment, as well as a relational database. Other services such as:

NoSQL databases,
Queue servers or
Managing External Users
are rare. If the new software is not exactly the client's main system, it is not possible to put it into production either. To do this, at least the following issues should be addressed:


However, the expertise required for new services is often lacking. The costs and risk facebook data of business disruption associated with the new service are often too high.

Managed services with cloud providers
For cloud providers, services are not costs and risks, but a product that generates money. As a result, investments are made differently here and services are suddenly available that could not be provided in the local data center.

In this blog post, I use Amazon Web Services (AWS) as an example. Google Cloud and Microsoft Azure offer similar services, but I am personally specialized in AWS.

In the cloud, we can run a service, such as a database, ourselves on a virtual machine. But this offers no advantage over our own data center. If we instead choose a database operated by AWS, AWS takes care of the operational issues mentioned above.


It doesn't always have to be SQL. AWS offers several NoSQL databases.

In the database space, AWS offers not only SQL databases , but also key-value, document, and graph databases , full-text search engines, and more exotic services such as blockchain databases . In total, AWS offers more than 200 services.

The impact on architecture
When we want to introduce a new service, such as a database, using a managed service from a cloud provider, we are building on an existing infrastructure with known strengths and weaknesses. This significantly reduces the risk of introducing the service. When introducing a self-powered service and integrating it into your own environment, there are always imponderables that can lead to delays and increased costs. When these risks are eliminated, one can better concentrate on the actual development of the application and dare to develop more innovative software. With the possibility of using new services apart from SQL databases and application servers, we can select and integrate the services that best fit our software. For example, in the past, due to the lack of alternatives, it was often necessary to

user administrations implemented,
which manages graphs of organizational structures in a relational database,
images and documents stored in a relational database and
simulated asynchronous events in a database.
With the right services, all of this is no longer necessary. The software is simplified, contains fewer errors and is finished faster.

Cost model
Consumption-based costs in the cloud require a different approach to operational costs.

On-Premise Data Center
In a classic software project, operating costs only play a role at the beginning of the project. The necessary hardware and licenses are then ordered. Once the software is up and running, operating costs have little impact. If changes are made to reduce license costs – for example, replacing an expensive database system with a cheaper one – this has little impact. Since the infrastructure is already in place and paid for in advance, more efficient software does not result in savings later on. Operating costs are not broken down into individual software components and, from a software development perspective, operating costs are considered to be existing costs.

In the cloud
In the cloud, there are different cost models based on consumption. More efficient software can result in immediate savings by generating fewer billable calls, using less memory, or requiring fewer or smaller virtual machines.