Microservice architecture : Part 1
Download as PPTX, PDF0 likes428 views
This document discusses the differences between monolithic and microservice architectures. In a monolithic architecture, all components of an application are interconnected and maintained within a single codebase, which can be inefficient to manage and scale as the codebase grows. Microservices address these issues by decomposing an application into smaller, independent services that communicate through APIs. This improves fault isolation, allows individual services to be developed and scaled independently, and makes it easier to adopt new technologies.
Microservice architecture : Part 1
- 2. Why? To understand why, we first need to have a look at what do we use. In general, it is a monolithic approach.
- 3. Monolithic Approach In simplest terms, Monolithic means composed all in one piece. In this approach, program components are interconnected and interdependent. In monolithic approach, we can divide the code in following layers: Presentation layer Business layer Data access layer
- 4. Monolithic Approach For example, when a web client (browser) sends a request, following happens: The business layer performs the business logic. Data access layer performs database collects/stores specific tasks. Presentation layer shows the data to the user. Basically, all code (presentation, business layer, and data access layer) is maintained within the same code base.
- 5. Problem 1 Due to one codebase, all changes need to be committed/deployed via single repository, which, as code grows, becomes very inefficient to manage.
- 6. Problem 2 Even in order to work in a certain module (say, product), developer needs to pull the whole codebase to his local and configure it which, in general, shouldn’t be the case.
- 7. Problem 3 Often, project tasks are divided based on the layer — UI team, backend team, database dev team etc. Teams are specialized in their domains, but they have little knowledge about other layers. So when there's a critical problem, it encompasses each layer and it takes additional time to decide which layer's problem it is and which team needs to address it. In case of a long running project, it may be possible that no single team understands the entirety of the application.
- 8. Problem 4 Due to large codebase, Scaling monolithic applications can often be a challenge. In a large, monolithic application, everything needs to be scaled together. For example, an application consists of two modules – user and product. Even if we want to scale product only, it is not possible in monolithic application.
- 9. Some more Adopting new technologies is cumbersome in monolithic applications since these applications are implemented using a single development stack (i.e. JEE or PHP etc), which, sometimes, limit the availability of the right solution for the problem. Monolithic applications are Fault intolerant so the application can crash, even by the failure of a single component/functionality.
- 10. Solution ?
- 11. Microservices A micro service is something which is self-contained, well-defined and does not depend on the state of other services. It is a small(er) independently deployable unit.
- 12. Transition to Microservices The idea is to split current application into a set of smaller, interconnected services instead of building a single monolithic application. Each microservice is a small application that has its own architecture consisting of business logic along with various adapters. Most microservices would expose a REST API and most services consume APIs provided by other services.
- 13. Technical impact of solution used It tackled the problem of complexity by decomposing application into a set of manageable services which are much faster to develop, and much easier to understand and maintain. Each service is independent so choosing programming language is more dependent to actual problem, rather than choices made at the start of the project which, ultimately, reduced barrier of adopting new technologies.
- 14. .. Continued It improved fault isolation so that the application can remain largely unaffected by the failure of a single service. Scaling services, as the need arises, is much cost effective. As per earlier example, an application consists of 2 modules – user and product. Now we can scale product only. Also, with the AWS Lambda like auto scalable services, it’s very easy and much more economical to scale micro-service, in case of increased demand for a particular service.
- 15. More coming soon ...