A raft consensus implementation that is simply and understandable
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
floyd become follower when request vote deined by some replic May 23, 2018
.gitignore Fix meta's voted_for_ip_port in ReplyRequestVote Mar 12, 2018
.gitmodules Update .gitmodules Nov 2, 2017
.travis.yml update .travis.yml Dec 26, 2017
LICENSE Initial commit May 27, 2016
README.md Update README.md Mar 13, 2018
README_CN.md update Chinese readme Aug 24, 2017
_config.yml Set theme jekyll-theme-tactile Jul 1, 2017

README.md

Floyd 中文

Build Status

Floyd is an C++ library based on Raft consensus protocol.

  • Raft is a consensus algorithm which is easy to understand;
  • Floyd is a library that could be easily embeded into users' application;
  • Floyd support consistency between cluster nodes by APIs like Read/Write/Delete;
  • Also some query and debug managment APIs: GetLeader/GetServerStatus/set_log_level
  • Floyd support lock operation upon raft consensus protocol

Users

  • Floyd has provided high available store for Meta cluster of Zeppelin , which is a huge distributed key-value storage.
  • Floyd lock interface has used in our production pika_hub
  • The list will goes on.

Why do we prefer a library to a service?

When we want to coordinate services, ZooKeeper is always a good choice.

  • But we have to maintain another service.
  • We must use its' SDK at the same time.

In our opion, a single service is much more simple than two services. As a result, an embeded library could be a better choice.

Floyd's Features and APIs

type API Status
Consensus Read support
Consensus Write support
Consensus Delete support
Local DirtyRead support
Local DirtyWrite support
Query GetLeader support
Query GetServerStatus support
Debug set_log_level support
  • Raft fetaures
Language Leader election + Log Replication Membership Changes Log Compaction
C++ Yes No No

Compile and Have a Try

  • Dependencies

    • gcc version 4.8+ to support C++11.
    • protobuf-devel
    • snappy-devel
    • bzip2-devel
    • zlib-devel
    • bzip2
    • submodules:        - Pink        - Slash
  • Get source code and submodules recursively.

git clone --recursive https://github.com/Qihoo360/floyd.git
  • Compile and get the libfloyd.a library
make

Example

Then right now there is three examples in the example directory, go and compile in the corresponding directory

example/simple/

contains many cases wrapped floyd into small example

Get all simple example will make command, then every example will start floyd with five node

make
  1. t is a single thread wirte tool to get performance
  2. t1 is multi thread program to get performance, in this case, all the writes is from the leader node
  3. t2 is an example test node join and leave
  4. t4 is an example used to see the message passing by each node in a stable situation
  5. t5 used to test single mode floyd, including starting a node and writing data
  6. t6 is the same as t1 except that all the writes is from the follower node
  7. t7 test write 3 node and then join the other 2 node case

example/redis/

raftis is a consensus server with 5 nodes and supporting redis protocol(get/set command). raftis is an example of building a consensus cluster with floyd(floyd is a simple implementation of raft protocol). It's very simple and intuitive. we can test raftis with redis-cli, benchmark with redis redis-benchmark tools.

compile raftis with command make and then start with run.sh

make && sh run.sh
#!/bin/sh
# start with five node
./output/bin/raftis "127.0.0.1:8901,127.0.0.1:8902,127.0.0.1:8903,127.0.0.1:8904,127.0.0.1:8905" "127.0.0.1" 8901 "./data1/" 6379 &
./output/bin/raftis "127.0.0.1:8901,127.0.0.1:8902,127.0.0.1:8903,127.0.0.1:8904,127.0.0.1:8905" "127.0.0.1" 8902 "./data2/" 6479 &
./output/bin/raftis "127.0.0.1:8901,127.0.0.1:8902,127.0.0.1:8903,127.0.0.1:8904,127.0.0.1:8905" "127.0.0.1" 8903 "./data3/" 6579 &
./output/bin/raftis "127.0.0.1:8901,127.0.0.1:8902,127.0.0.1:8903,127.0.0.1:8904,127.0.0.1:8905" "127.0.0.1" 8904 "./data4/" 6679 &
./output/bin/raftis "127.0.0.1:8901,127.0.0.1:8902,127.0.0.1:8903,127.0.0.1:8904,127.0.0.1:8905" "127.0.0.1" 8905 "./data5/" 6779 &
└─[$] ./src/redis-benchmark -t set -n 1000000 -r 100000000 -c 20
====== SET ======
  1000000 requests completed in 219.76 seconds
  20 parallel clients
  3 bytes payload
  keep alive: 1

0.00% <= 2 milliseconds
0.00% <= 3 milliseconds
8.72% <= 4 milliseconds
95.39% <= 5 milliseconds
95.96% <= 6 milliseconds
99.21% <= 7 milliseconds
99.66% <= 8 milliseconds
99.97% <= 9 milliseconds
99.97% <= 11 milliseconds
99.97% <= 12 milliseconds
99.97% <= 14 milliseconds
99.97% <= 15 milliseconds
99.99% <= 16 milliseconds
99.99% <= 17 milliseconds
99.99% <= 18 milliseconds
99.99% <= 19 milliseconds
99.99% <= 26 milliseconds
99.99% <= 27 milliseconds
100.00% <= 28 milliseconds
100.00% <= 29 milliseconds
100.00% <= 30 milliseconds
100.00% <= 61 milliseconds
100.00% <= 62 milliseconds
100.00% <= 63 milliseconds
100.00% <= 63 milliseconds
4550.31 requests per second

example/kv/

A simple consensus kv example contain server and client builded with floyd

Test

floyd has pass the jepsen test, you can get the test case here jepsen

Documents

Contant us

Anyone who is interested in raft protocol, used floyd in your production or has wrote some article about souce code of floyd please contact me, we have a article list.