MicroservicesDesignPatterns.txt

				Microservices
...............................................................................................
Distributed Enterprise application:
   The application developed, tested, deployed on server , distributed over network (internet)

Patterns(ways) for building Distributed Enterprise application:

Before 2010  -Monolithic Enterprise app
After 2010  - Microservices Enterprise app

Enterprise - domain - Banking

Banking:
  Modules
   1.Account Management - Service (mini application)
   2.Customer Management
   3.Loans Management


Microservice is way of building app testing, deploying, maintaining, independently.

Objective:
   Automation. -  Dev Ops and Cloud
   CI AND CD

Deployment:
   Moving app on to the data center.

Data center:
  Servers + os(Linux)----> bare deployment...
  VM -oracle vm
  Lightweight vm - Containers - Docker

Cloud Computing:
   Renting data centers


Application(software system) Development patterns:
.................................................

Network based Applications - Distributed Application

Application has layers:

1.User interface layer
2.Application biz layer
3.Data Layer /Repository layer

History of Architecture of Distributed Application

Distributed means, application is broken into multiple parts and each part put into multiple hosts/machines, connect those via network.

1.Mainframework based distributed
   1.Application biz layer
   2.Data Layer /Repository layer
 Where as User interface layer is kept in dump terminals connected to mainfram.

Drawbacks:
1.Too costly
2.scalability is too difficult.


Advantage:
1.High security
2.Centeralized management.

2.Client Server Architecture

2.1.Main framework based client -server , where as mainframe acts as server and digital computers act as clients.
 
2.2.Digital computer based client - server Architecture
   Servers and clients are digital computers

Based on this we can classify the applications  layered /tiered concept

1.single tier/layer
   client, server, database every is kept in one single machine...
2.two tier/layer
  user interface is kept in client machine,
  data logic and biz logic is kept in server machine
 both machines are connected via networks

          "This arch based on LAN /WAN"

3.three tier /layer

 This arch based on "internet network" and web computing

   client - client machine
   server - biz logic is kept inside another machine
   database - is kept inside another machine

 Client is browser
 Server BIZ logic is kept as "web Applications"
 Database is accessed by "Server side technologies - J2EE,ASP/.net,PHP,....

4.N-tier / layer

 Client is browser
 Server BIZ logic is kept as "web Applications"
   -Again spilt into multi layered
 Database is accessed by "Server side technologies - J2EE,ASP/.net,PHP,....
 In 2000 , J2EE introduced n-tier client server  

  browser -------web application(servlets/jsp)----EJB----Messaging/Databases(JMS/JDBC/Middlewares)

 Spring based N-tier client server arch:

  browser -------web application(spring mvc)---Spring services----Spring data----Messaging/Databases(JMS/JDBC/Middlewares)
.....................................................................................
		    How to build N-tier distributed Applications
.....................................................................................

Steps/Process:

1.Domain Modeling

  Banking, Online Food Delivery App, Ecommerce Domain

2.Select technology

   if your app is based on web and internet.

  Steps:
   1. Database -  Oracle 
   2. MOM - RabbitMQ,IBM MQ,Microsoft MQ
   3. Development Technology
       Java/JEE  - Why you go with specific implementation technologies?
       .Net
       Php

3.Development and release methodology
   Waterfall -  traditional dev , release


Any domain consist of various modules
 -Accounts
 -Loans
 -Customers
 -Card
etc.....
 
4.Testing
   Once the development is over, the app is going to be under testing

5.Production
   Once the app is tested fully, ready for production.

6.Maintance
   Once the app in the production, it goes on maintaince...

if any app is built based on the above methodology, which is called as "Monolithic"
.....................................................................................
 Challenges in the application development ,testing, release, Production ,maintaince
....................................................................................


1.Every thing has to go step by step -  this increase cost , time waste, resource waste

Companies like Amazon, Netflix who wanted fast development,test,release,maintaince : Dynamic  methodology to build applications -  No downtime,
	One module takes more time ,another module takes less time, because of one module , other module should not wait.

2.Technology bottleneck - Mono technology
 
 The whole application is built using single technology - Java - vendor lock
 The whole application targets single database - Oracle /Mysql/Microsoft SQL server..

3.Employing security layer is more complicated

4.Deployment / Production.

   The dev and prod env is completely different
   Bare deployment models
   VM based deployment...

...................................................................................
								  New way of building apps


1.Automatation is key concept 

   from the  Analysis ,dev ,test ,release, prod, maintaince

Agile :(Requirement Analysis)

  Agile is an iterative approach to project management and software development that helps teams deliver value to their customers faster and with fewer headaches. Instead of betting everything on a "big bang" launch, an agile team delivers work in small, but consumable, increments

Breaks the application into smaller and smaller.
 - fast delivery with quality on time.


Requirements are highly dynamic, cant be free zed, since it is dynamic start development ,test, release, deploy  periodically.


We need automation, through which  automatically only we can achieve fast delivery -
 in order to automate, a new technology was created "Dev Ops" - Dev + Operations together.

Distributed source code repo - git

Pipe lines tools  - 
   Jenkins -(Continuous Integration)

Requirement---> Dev---push the code to source code repo---|CI Tool---Compile--Build/pack--Testing-Deployment(CD)

Every thing is here Continuous happens

Continuous Req Analysis
Continuous Dev 
Continuous release /build
Continuous test
Continuous release /build
Continuous deployment
Continuous tracing and monitoring

This process applied on every module in the applications

Order Management 
   Continuous Req Analysis,Dev,release,test,deployment,tracing,monitoring

CustomerManagement 
   Continuous Req Analysis,Dev,release,test,deployment,tracing,monitoring


if any app is built based on the above methodology, that application is called as   "Microservice"
....................................................................................

What is Microservice?
  Microservices - also known as the microservice architecture - is an architectural style that structures an application as a collection of two or more services that are:
1.Independently deployable
2.Loosely coupled

Services are typically organized around business capabilities. Each service is often owned by a single, small team.

A Service is a "mini application" that  implements narrowly focused functionality.
Such as orderManagement, customer Management, and so on....

....................................................................................
							How to design and implement microservices

The microservices is all about practices followed, implemented, and tested in real time production grade applications in various companies like amazon,netflix,google,microsoft.

The many community people joined  together who formed the pattern language in order to begin development of Microservices - Microservice pattern language, design patterns

if i want to build microservice based apps, how to begin , where to begin.

			Microservices design patterns
.....................................................................................
....................................................................................
		   Microservice architecture pattern language
...................................................................................  
The Microservice pattern language is a collection of patterns that help you architect an application using the microservice architectures.

Infrastructure Patterns:
  Theses solves problems that are mostly infrastructure issues outside of development.

Application patterns:
  These are for related to development

Application Infrastructure:
   Application related infrastructures like containers
....................................................................................
					     Patterns for Decomposing an Application into services	


1.Decompose by business capability 
	 |
	 |
2.Decompose by subdomain

3.SelfContained Service

4.Service Per Team
.....................................................................................
			    Design patterns in Microservices
....................................................................................

Application Architecture Pattern
   For building n-tier client server distributed application.

-Monolithic architecture
-Microservice architecture

Decomposition Pattern -Y scalling

1.Decompose by business capability 	 
2.Decompose by subdomain
3.SelfContained Service
4.Service Per Team


MicroService Architecture Pattern------>Depcompostion Pattern

Decompose by business capability 

If you are going to build online store.

Business capability:
Product Catalog Management
Inventory Management
Order Management
Delivery Management.

Alternate Pattern

Decompose by SubDomain:
 Decompose the problem based on DDD principles.

....................................................................................
....................................................................................
				Data Management
....................................................................................

Core Pattern:
1.Database Per Service Pattern
2.Shared Database

Note:
  if you take any data related patterns, "Transactions" are very important.


1.Database Per Service Pattern leads/succeeds other patterns
  Domain Event
  Event Sourcing 
  Saga - Transaction
  CQRS
  API Composition
 
..................................................................................
	    Advance Data Management Pattern -Transactional Messaging Pattern   
..................................................................................

1.Transactional outbox
    2.Transactional log tailing
     or
    3.Polling publisher

   2.1.Idemponent Consumer
.....................................................................................
.....................................................................................
			Communication Style Patterns
...................................................................................

Service = Mini Application

MiniApplication = Collection of Programs

Collection of programs in Java = Collections of classes

Collections of classes = Collections Objects.

Object/Class = Collection of  state and behaviour

State = data
Behaviour=methods

Object = methods

methods = API

API will DO three things

1.write - update,remove,insert
2.read
3.process

class OrderService {

   @Autowrited
   private OrderRepository orderRepo;
   //API
   public List<Order> findAll(){
	orderRepo.findAll()
   }
 
}

Types of API:
1.local api
   api which are called with in same runtime by other apis
2.remote api
  api which are called outside runtime via networks 

How to build remote api?

 Based on protocals

1.HTTP Protocal.

 if you design your api based on HTTP protcal, those apis are called as "WebServices"

Web Service:
  RESTFull WebServices,SOAP WebServices

REST API = Program

In java => classes

In Web Services classes are called "End Points"

In Micro services -Services can be  represented as "WebServices"

Rest WebService------>http-----RestWebservice => HTTP based Microservice

Rest WebService------>http-----Graphql => HTTP based Microservice

Rest WebService------>http/2 over tcp------GRpc Service=> TCP based Microservice

Rest Web Service ---->TCP/MOM-------------->Messaging Service -Middlewares

Communication Sytle patterns:

1.RPI patterns
   REST,gRPC,Apache Thrift - RPI implementations
2.Messaging
   Any Messaging middlewares - RabbitMQ,IBM MQ,MicroSoft MQ - MQTT,AMQP
   Streaming platforms - Apache Kafka,Confluent Kafka
   2.1.Idemponent Consumer
3.Domain Specfic Protocal
   SMTP - Mail Service   

.....................................................................................
			   Deployment Patterns
.....................................................................................

Once the services(applications) are ready, we can move the application into production.

Production Related Patterns:

Deployment Environment/Plattforms

1.Bare Metal 
    Where as physical hardware, and operating system, Where we can provision our application.
    If you deploy java application.

  OS: Linux
  JRE- 17
  WebContainer -Tomcat
  Databases -MySql
  Streaming Platforms-Kafka
   
2.Virutal Machine 
   Oracle Virtual box 
    on VM , you can install os-linux
  JRE- 17
   WebContainer -Tomcat
   Databases -MySql
   Streaming Platforms-Kafka

3.Containerized Deployment
    It is lightweight vm - Docker and Kubernets
   JRE- 17
   WebContainer -Tomcat
   Databases -MySql
   Streaming Platforms-Kafka
  
4.Cloud 
   ->VM /container/bare 
  you can just deploy your app only,
  cloud may give you all softwares for you...


  "Cloud with containers are most preferable deployment for microservices"

Design patterns:

Bare Metal:
 1.Multiple services instances per host
 2.Service instance per host
VM
 1.Service instance Per VM
Container
 1.Service instance per Container
Cloud
 1.server less deployment
 2.Service deployment platform
 3.container and cloud

			      
		if your app deployement is in container 
				  or 
    			       in cloud
				  or 
			   Container with cloud
				 or 
			   in any Virtualized Env

if any micro service(application) is running in containerized env like kubernets(docker).

Challanges:
 1.suppose the application is accessed by other application or external application
   we need to communicate the application with help of "host:port".
 if application is running Virtualized env, "host and port" is not static,it would be dynamic.

 if it is dynamic then how other microservices, and external application, how they can communicate.
			
To solve the problem of identifying the services which are running in Virtualized env

Advanced Communication Patterns
				(Service Registry and Discovery)


When we apply this pattern, services never communicate "directly", because they dont know each other due to "dynamic location",so they use broker to communicate, broker will have all service information-Service Registry

Service Registry Patterns:

1.Client side service Discovery
2.Server-side service Discovery
   ->Service Registry
  	->Self Registration
   	->ThridParty Registration

.....................................................................................
.....................................................................................
                     Services are running in Virtualized Env
		     Services are talking via Service Registry
		What if i any service is down / Slow / Throwing Exception
					
Microservices provides a Design patterns to handle failures and slow calls

			 Service Reliablity Patterns

1.Timeout Pattern
2.Bulk Head Pattern
3.Retry Pattern
4.Circuit Breaker Patterns
.....................................................................................
.....................................................................................
			    Configuration Data and Its patterns

Every application which requires configuration data,the configuration data may be connection strings,api tokens,application settings etc...

In Java application, configuration data is kept inside properties or yml files...

What if in micro serivces, the configuration is need to be shared across the application?

We have design pattern to centeralize configuration data/information.

1.Microservice Chassis
2.Service Templates
3.Externalized COnfiguration

.....................................................................................
		     Micro services are ready in production
		We have exposed our microservices via API Gateways
			  How to secure them?

Security Patterns

1.Access Tokens
   -Authentecation
   -Authrozation
   -SSL
   -Policies

.....................................................................................
		   Micro services are ready in production
			Now we need to expose to
		 other Applications- User interface applications

Microservices provide you a design pattern, called

External API patterns:

1.API Gate ways
  2.Back End for FrontEnd

.....................................................................................

.....................................................................................
....................................................................................
			Now your Micro service is in Production
				Next what should i do
			     Your App in Maintaince 

Monitor Your apps......

Observablity Design patterns:

1.Log Management/Aggregation Pattern
2.Application Metrics Pattern
3.Audit Logging pattern
4.Distributed Tracing
5.Exception Tracker Pattern
6.Health Check API pattern
................................................................................

..............................................................................................
How to build Microservices?
  There are plenty of  Patterns available 

How to implement these patterns and where to implement?

 All programming languages support microservices via frameworks and libs
***********************************************************************************************************************************************************
Microservices implementation:

Java frameworks:
 1.Spring boot with Spring cloud
 2.Quarkus
 3.Vertx
 etc.....

JavaScript:
 1.node.js with "Molculer.js"
 2.node.js with "nest.js"
 3.nodejs with "loopback.js"
 etc.....