3

What I know is:

  • Consistent hashing: uniform distributed storage system
  • Cone hashing: non uniform distributed storage system

I want to know:

  • How it works?
  • What is the use of it?
  • What is the difference between this two types of hashing?

I am not able to understand the difference between this two. Please someone help me with this!

Basit Anwer
  • 6,742
  • 7
  • 45
  • 88
Hemanth Venkatappa
  • 39
  • 1
  • 1
  • 8
  • 1
    Have you done any research at all? A Google search on "cone hashing" reveals a lot of information. But you'll have to put forth a little effort and actually *read* the papers you get from the search. It looks like cone hashing is an alternative to consistent hashing for distributed systems. – Jim Mischel Apr 15 '15 at 13:43
  • 1
    I did a google search but I am not able to find a solution for this. If I find also then it is not clear and difficult to understand. I appreciate if there is some simple answer for this. – Hemanth Venkatappa Apr 18 '15 at 11:14

1 Answers1

2

The simple difference is in its definition.

Consistent Hashing is based on division of keys based on the number of datastores it needs to be distributed to.

e.g. 1000keys to be distributed to 5 nodes means 250 keys to each node.

Cone hashing is different in a way that it distributes the keys based on an algorithm which assigns keys based on available resources rather than distribution. Node A has more storage therefore most of the keys will be distributed to it e.t.c

Simple use cases would be IMDGs or distributed caches, whereas for cone it would be where the machines are of different specs so you could distribute keys based on their specs.

Basit Anwer
  • 6,742
  • 7
  • 45
  • 88