Checkout this concepts

[pratikGhodke1]

• [x] Api Throttling
• [x] Microservices
• [x] Service registry and discovery and tools

[As a task you can add notes here in comments]

[karelumair]

API Throttling (Rate Limiting)

• Throttling basically means controlling the flow of something.
• The API throttling is the process of limiting the number of API requests a user can make in a certain period of time.
• This is an important parameter to consider while designing a resilient system or an API.
• Each unique client or device gets a rate limit for making API requests, once this limit is exceeded no more requests are accepted by that particular client for a specific period of time.
• This is used to balance the load or traffic on the system. Also, it helps to protect the system against malicious attacks.

Algorithms used for Throttling:

• Fixed Window
• Sliding Window
• Leaky Bucket

[karelumair]

Microservices

• Microservices or Microservice architecture is basically an architectural style of structuring an application as a collection of services.
• This approach enables rapid and reliable delivery of large, complex applications.
• The main idea behind this approach is to break down a large monolithic application to have each independent services working on its own which helps in rapid development and scalability.
• These services communicate via a well-defined interface using lightweight APIs.

Benefits:

• Scalability
• Easy deployment
• Resilient

[karelumair]

Service Discovery and Registry

"Service discovery is basically a technology to automatically detect services and devices on a network."

• In microservices architecture there are multiple services deployed on different servers.
• The key challenge here is How does one service know where the other service is depolyed?.
• One solution is to hard code the urls, but this is not a good idea as url might change later.
• This is where service discovery is used.

Service Discovery

• The idea is to have a server called discovery server, Each service in the architecture of application registers itself to the discovery server.
• The discovery server acts as the service registry, storing the services and their urls (information).
• This process of asking a discovery server to get to a service is called a service discovery.

Two types of service discovery.

Client Side

• In this, Client makes first call to the service registry to find the service needed.
• Then the client makes second call to the actual service discovered.
• Disadvantage here is that, client service has to make two calls.
• Ex. Eureka Server, etc.

Server Side

• In this, Client makes request to service registry.
• Then, the service registry transfers the request or redirect it to the required service.
• Advantage here is that, client doesn't need to make multiple calls.
• Disadvantage of this approach is that, discovery server should know how to do all the transferring stuff and must be running all time.
• Ex. NGINX, AWS, etc.

Service Discovery Tools

• Docker
• AWS Cloud Map
• ZooKeeper
• Eureka
• Etcd etc....

[pratikGhodke1]

Can you also check -

• [x] What are the options/frameworks available to implement microservice architecture in python.
• [x] Advantages and disadvantages to implement it in python
• [x] Better solutions that python (gonna be java :) )

[karelumair]

Frameworks Available for microservice architecture in python.

• Fast API
• Flask
• Torando
• Falcom
• Bottle
• nameko

[pratikGhodke1]

You can also take a look at nameko framework which works on asynchronous queue-based mechanism.

[karelumair]

Advantages and disadvantages of implementing microservices in python

Advantages

• Python is dynamic, interpreted language which makes it easier for writing small scripts.
• Python is also a dynamically typed language, hence doesn't require to use variable type while defining variables.
• Python is compatible with legacy languages such as PHP and the ASP.NET framework.
• Python's supports various libraries for writing REST services, and Python community is large.

Disadvantages

• speed of execution and the nature of the interpreter environment are the biggest potential drawbacks of python to develop microservices.
• since python is interpreted language, this leads to slower execution time as compared to other compiled languages.
• since python is dynamically typed, it may lead to error at the time of execution which were ignored during compilation stage. This may lead to the crashing of application.

[pratikGhodke1]

Python actually is not a strongly typed language. Its dynamically type and any variable can be re-assigned to contain different data type without explicit data type conversion. It does have some data type checks and enforcements though.

[karelumair]

oh yes. Actually I was also confused coz I knew it is dynamic but that article refferred as strongly typed. I searched it and it said Python is both strongly typed and dynamically typed. That's why mentioned it.

[karelumair]

Better solutions than python

Java

• Java's Annotation syntax, forms the key factor that makes it a great programming language for developing microservices.
• It offers stability and consistency for distributed applications.
• Java EE standards are well suited for microservices applications.
• Java Spring Boot framework uses auto-configuration tool to quickly create REST APIs and link applications.

Golang

• Go is a statically typed and compiled language, having strong memory safety and garbage collection as some of its main features.
• It offers a safe and straightforward way to manage complex web service instances and distributed collections of functional libraries.
• Go is also highly concurrent which forms a major advantage for developing microservices.
• Framework such as GoMicro can be used.

.NET

• .Net framework makes it simple to build the APIs that becomes the microservices.
• It includes built-in support for building and deploying microservices using Docker containers.
• It is possible to mix the .Net microservices with applications written in Java, Node JS, or any other languages.
• the .Net microservices can run on all leading cloud platforms.

[karelumair]

Why language choice matters in microservices development?

• since microservices are designed to operate as independently as possible, they must still communicate and share data using various techniques.
• Rather than communication via centralized messaging system, they must communicate between themselves to perform tasks and to scale as needed.
• This is where language choice is important.
• The language choosen for the microservices based application should ideally support a number of important communication formats and protocols.
• Any language used to create microservices should support REST, which primarily relies on HTTP requests that post, read and delete data.