Dec 31, 2021
System Design 1.0
Well, it’s gonna start from the beginning and go up to interview-level design thinking.
if the client wants to talk in HTTP protocol, as port we use 80, if it’s HTTPS use 443. Those are decided a long time ago. But good to keep that sound knowledge.
ports..Let's talk through a port. I’m gonna use nc -l 8081 command.
What it basically does is the in-network manner, one of my terminal gonna listen to that port. Keep looking only on that particular window with hope :)
Let’s not disappoint that guy, Since we are on the same machine, let’s talk with our local Ip.
Yes, it‘s basic (No worries, gonna master these :).Keep in mind that ports 0 to 1023 are reserved, which are called well-known ports.
Better to have a clear idea bout TCP.TCP helps to make sure the delivery of a set of IP packets without corruption.
DNS
Short for Domain Name System, it describes the entities and protocols involved in the translation from domain names to IP Addresses. Typically, machines make a DNS query to a well-known entity that is responsible for returning the IP address (or multiple ones) of the requested domain name in the response.
Better to have a clear idea about Web socket, find it here:
When system designing Service provider gives us the SLA and SLO.Simply availability in a year and objective in service level.
Some cloud infrastructure provides 99.999% availability, which means 5 nine, cool.
But How they keep those available? They follow active redundancy and passive redundancy. Passive Redundancy means, when 1 of 5 servers is down, the other 4 take care of the load. To understand active redundancy imagine there are few machines, but only one or two machines handle the load, when those real workers get down, others somehow get to know about that,take over the load .Yeah kind of ice break time:)
But keep in mind that, some parts of the systems are bit okay to fail, yeah,, just admit that :) , means not the end the whole future of the company.
We can multiply some components of the system, to avoid the single point of failure, oh forgot that’s called redundancy. While we are creating a system we have to care about hr latency and throughput too. Throughput is determined by the cloud service providers if we use a cloud platform, we can increase the amount. Basically, it’s talking about how much data can carry on.
Many things to come in this reading list …
