All About HTTP (Hyper Text Transfer Protocol)

HTTP Anatomy

It’s Client/Server Architecture. Client (Browser) sends HTTP Request to the Server and then Sever respond back to the Client with HTTP Response.

• (Client) Browser, Python or JavaScript app, or any app that makes HTTP Request.
• (Server) HTTP Web Server, eg. IIS (Internet Information Services), Apache Tomcat, NodeJS, Python Tornado.

HTTP Request

Format	Description
URL	The URL which Locate to HTTP Server
Method Type	POST, GET, PUT, DELETE
Headers	The context behind the payload like Cookies, Proxy etc…
Body	Actual Payload (GET won’t have body…)

HTTP Response

Format	Description
Status Code	Status of Operation (200 - Success, 404 - Not Found, 400 - Bad Request)
Headers	The context behind the payload
Body	Actual Response Data (Payload)

How HTTP works?

• HTTP is Layer 7 Protocol that means it operates within the application layer of the Open Systems Interconnect (OSI) model.
• Like TCP is Layer 4 Protocol which is actually smart layer which transfer payload (blob data) by establishing TCP connection from source to destination and transfer data in terms of packets.

• TCP is little bit expensive because it needs to do lots of stuff like congestion control (traffic management), security stuff (cryptography), etc…

• Before getting request and sending request there is handshaking take place to establish secure connection between Client and Server.

HTTP 1.0

• It was launched in 1996. at that time max capacity of RAM is 64 MB or around.

• TCP connection was expansive at that time. for each request we open connection with handshaking, then request-response takes place and close the connection.
• If we want to fire another request we again need to open that expansive TCP connection and request-response and close the connection. It’s tedious & slow process to work with.
• So, HTTP 1.0 didn’t sustain more than a year even.

HTTP 1.1

• In 1997, HTTP 1.1 was launched. It is the only protocol that survive around 20 Years. then HTTP 2.0 takes place.

• It solves a major issue which was there in HTTP 1.0 which was it needs to establish the connection for each request.
• In HTTP 1.1 It solve the issue by persisting that connection between Client & Server.
• For Persistent It uses caches, etag. That ultimately helps to reduce down latency.
• It also provide streaming with chunked transfer.

HTTP 2

HTTP 3