• Flattening of data structure as much as possible.
• Storing entities separately.

relationships with unique IDs.

Bad approach:

student{
	name: "siivaji",
	City: "Amalapuram ",
	College:{
		Id:"cg1",
		name:"aditya College of Engineering ",
		Pincode :5454643,
	}
}

Good approach:

student {
	name: "siivaji",
	City: "Amalapuram ",
	College: cg-id
}
College {
	Id:"cg1",
	name:"aditya College of Engineering ",
	Pincode :5454643,
}

Why?

• Remove redundancy.
• efficiency.
• Simplifies Nested relationships.

without Normalization :

const state = {
  users: [
    {
      id: 65577,
      name: "shivaji897",
      posts: [
        {
          id: 101,
          title: "post1",
        },
        {
          id: 102,
          title: "post2",
        },
      ],
    },
    {
      id: 65577,
      name: "vyshnavi34",
      posts: [
        {
          id: 101,
          title: "post1",
        },
        {
          id: 102,
          title: "post2",
        },
      ],
    },
  ],
};

console.log(state.users[1].name);
console.log(state.users[1].posts[1].title);

normalized data :

const normalizedState = {
  users: {
    byIds: {
      1: {
        id: 65577,
        name: "shivaji897",
      },
      2: {
        id: 65577,
        name: "vyshnavi34",
      },
    },
    allIds: [1, 2, 3, 4],
  },
  posts: {
    byIds: {
      101: {
        id: 101,
        title: "post1",
        userId: 1232,
      },
      102: {
        id: 102,
        title: "post2",
        userId: 2556,
      },
      103: {
        id: 103,
        title: "post3",
        userId: 1656,
      },
    },
    allIds: [101, 102, 103],
  },
};

// O(1) complexity and no redundancy, objects.keys don't give the proper order(random).