Skip to content

Latest commit

 

History

History
72 lines (46 loc) · 2.87 KB

README.md

File metadata and controls

72 lines (46 loc) · 2.87 KB
Raw

Raft-Simulation-with-Go-Routines

It is a simple simulation of raft protocol leader selection algorithm with Go routines and channels

Installation:

Pre-requsites :-

  • Install Golang (v1.19)
  • Install Git

Run this command in terminal :-

git clone https://github.com/Teja2045/Raft-Simulation-with-Go-Routines

Run:

Run this command to run the simulation :-

go run raft.go

Features:

  • Leader Election
  • Failure Detection and Recovery
  • Single Leader at a time
  • simple state machine
  • Data communication via go Channels

Implentaion Details

  • Considered 5 nodes for simplicity (increasing the number of nodes won't cause any problem).
  • At the start, Each node is initialized with ID, Term=0, different timeouts like Election timeout, Heart Beat timout and different channels like voteRequest channel, HeartBeat Channel for different type of communications.
  • They also store an array of peer nodes along with their details (though, just storing channels information is enough as it is implemented in such a way to handle abstraction).
  • Each node will start running in an Infinite Go routine.
  • The State Machine involves three states for Each Node, FOLLOWR state, CANDIDATE state and LEADER state
  • Intially all the nodes starts as FOLLOWERS.
  • But as the time goes on, they will be on time out and becomes CANDIDATE.
  • By increasing their term, the node that became CANDIDATE will ask for Vote Request (term + its id + its own voting channel) to all its peers via their vote Request Channel.
  • The nodes that recieve vote request will handle the request depending their own state and term and send the response (vote/reject) via the response vote channel of requester. The response channel is embedded in the request itself, so they dont need to look their peer information for it!
  • If a CANDIDATE receives more than 1/2 number of nodes of votes, it will become a LEADER. otherwise, it either waits for Election timout or pings from other nodes with higher terms.

Note:

  • Note-1: Replicated Log is not implemented

  • Note-2: Might exhibit weird behavior as routine's runtime is uncontrollable

  • Note-3: changing timeouts should be handled crefully as algorithm most dependent on them

  • Note-4: this is an implementation to my understanding and simplified to be a fit for Go routines and channels

      Note: any suggestions and advices are welcomed! Please make a pull request or create an issue to discuss more!

To run through a docker image:

  • Clone this repository.
  • cd to the directory containing the Dockerfile.
  • Build the docker image using the following command:
    docker build -t raft_simulation_with_go_routines .
  • Run the following command to the start the built image in the previous step:
  docker run -it --name raft_simulation_container raft_simulation_with_go_routines

Thank you!