Giter VIP home page Giter VIP logo

asio_kcp's Introduction

asio_kcp - a fast and reliable udp framework.

####This framework will be the fastest protocal for pvp game.

####kcp - a A Fast and Reliable ARQ Protocol (Can use with UDP) #####website of kcp: https://github.com/skywind3000/kcp #####This project wrap kcp by boost asio. Making it easy to use.

Feature

  • Using in realtime pvp game that need miniest network delay. For example, the multiplayer first person shooting game.
  • Reliable UDP protocal.
  • Minimize delay. And the worst delay should be good. 3 or more times better than enet when worst network lag happen.
  • Has good performace in wifi and phone network(3G, 4G).
  • Extra using 20% ~ 50% network flow for speed improvement.

Future Feature

  • Encode the communication for encryption and decreasing the net through.

###Bench mark ``` worst network lag happen: asio: 10:51.21 291 295 269 268 231 195 249 230 225 204

enet: 10:51.21 1563 1520 1470 1482 1438 1454 1412 1637 1588 1540


#####You can see the bench mark test result at [here](https://github.com/libinzhangyuan/reliable_udp_bench_mark)
#####Or you can see the bench mark detail log at bench_mark folder.




### client_lib
#### Feature:
* Do not using boost. And do not using feature of c++11
* It's implement by original thread and socket.
* [client_lib_usage](./client_lib_usage.md)



<br><br>
### You can use Dockerfile to create a [docker](www.docker.com) container for compiling or create compiling env your self.<br>

<br><br>
### Using compiling env in [docker](www.docker.com)<br>
* This is simple if you familar with docker. Please read the text at [docker/docker_readme.md](./docker/docker_readme.md)



<br><br>
### Create compiling env yourself<br>
* I using gcc 4.8
* Other version of gcc that support c++11 should be OK. You can compile asio_kcp as verifying.

* I using boost 1.58
    * Other version of boost should be OK. You can compile asio_kcp as verifying.

* Build g2log
    * download g2log from [official website](http://www.codeproject.com/Articles/288827/g-log-An-efficient-asynchronous-logger-using-Cplus#TOC_initialization)
    * -  or using the package in third_party folder.
    * cope g2log folder to third_party folder
    * $ cd third_party/g2log/ &&  mkdir build && cd build && cmake .. && make

* Build muduo
    * $ cd third_party/
    * $ git clone https://github.com/chenshuo/muduo.git
    * -  or using the package in third_party folder. third_party folder.
    * $ cd muduo
    * compile on OSX:  $ patch -p1 < MacOSX.diff
        * edit muduo/CMakeLists.txt.
            * find set(CXX_FLAGS  then add 2 new line:
                * -D__GXX_EXPERIMENTAL_CXX0X__
                * -std=c++11
            * comment -march=native
            * comment set(CMAKE_CXX_COMPILER "clang++")
            * comment -Werror
            * uncomment -DMUDUO_STD_STRING
    * compile on linux:
        * edit muduo/CMakeLists.txt.
            * find set(CXX_FLAGS  then add 2 new line:
                * -D__GXX_EXPERIMENTAL_CXX0X__
                * -std=c++11
            * uncomment -DMUDUO_STD_STRING
            * comment -Werror
    * modify muduo/base/LogStream.h   kSmallBuffer = 4000  ->  kSmallBuffer = 4000*4
    * adding VERBOSE=1 to "make" in muduo/build.sh will show detail of compiling.
    * $ CC=gcc CXX=g++ BUILD_DIR=./build BUILD_TYPE=release BUILD_NO_EXAMPLES=1 . ./build.sh

* modify the BOOST_LIB_PATH and BOOST_INC_PATH in allmake.sh
* do compiling at project root folder: $ . allmake.sh

The third_party folder should like below,

▾ third_party/ ▾ g2log/ ▸ build/ ▸ src/ ▸ test_example/ ▸ test_performance/ ▸ test_unit/ CMakeLists.txt ▸ muduo/


### Compile unit test
* download gtest from [website](https://googletest.googlecode.com/files/gtest-1.7.0.zip)
    * or just use the package in third_party folder.
* unpack the gtest-1.7.0.zip. coping gtest-1.7.0 folder to asio_kcp/third_party/
* $ cd gtest-1.7.0 && ./configure && make
* download googlemock from https://googlemock.googlecode.com/files/gmock-1.7.0.zip
    * or just use the package in third_party folder.
* unpack the gmock-1.7.0.zip. coping gmock-1.7.0 folder to asio_kcp/third_party/
* $ cd gmock-1.7.0 && ./configure && make
* back to asio_kcp folder then $ . utest_make.sh
* $ ./asio_kcp_utest/asio_kcp_utest

The third_party folder should like below,

▾ third_party/ ▸ g2log/ ▸ gmock-1.7.0/ ▸ gtest-1.7.0/ ▸ muduo/



###Run example test
##### filter the verbose log from asio timer
    ./server/server 0.0.0.0 12345 2>&1 | grep --line-buffered -v -e deadline_timer -e "ec=system:0$" -e "|$" >>bserver.txt
##### filter all asio log
    ./client_with_asio/client_with_asio 23425 127.0.0.1 12345 500 2>/dev/null




### run bench mark test of kcp.
* git checkout kcp_bench_mark_test
* compile
* then run server on your server:
    * ./server/server 0.0.0.0 12345 2>&1 | grep --line-buffered -v -e deadline_timer -e "ec=system:0$" -e "|$" >>bserver.txt
* if you want to test the 3G/4G. you can share the wifi on your phone by using wiless AP. Making your client computer connect to this wifi.
* run client on your client computer (Note: changing the ip and port to your server)
    * ./client_with_asio/client_with_asio 23445 120.26.200.117 12345 500 2>/dev/null
       Note: changing the ip and port to your server which is running the asio_kcp_server


### run bench mark test of kcp in docker container.
* git checkout kcp_bench_mark_test
* sudo docker run -it --name asio_kcp_server -p 12345:12345/udp -v /home/zhangyuan/work/asio_kcp:/home/work/asio_kcp asio_kcp:develop
* compile
* then run server on your server:
    *  cd /home/work/asio_kcp
    *  ./server/server 0.0.0.0 12345 2>&1 | grep --line-buffered -v -e deadline_timer -e "ec=system:0$" -e "|$" >>bserver.txt


sudo docker run -it --name asio_kcp_client --net=container:asio_kcp_server -v /home/zhangyuan/work/asio_kcp:/home/work/asio_kcp asio_kcp:develop
* if you want to test the 3G/4G. you can share the wifi on your phone by using wiless AP. Making your client computer connect to this wifi.
* run client on your client computer (Note: changing the ip and port to your server)
    *  cd /home/work/asio_kcp
    * ./client_with_asio/client_with_asio 23445 127.0.0.1 12345 500 2>/dev/null
       Note: changing the ip and port to your server which is running the asio_kcp_server


<br>
###Centos 6.5 install help:
#### boost  http://www.boost.org/doc/libs/1_58_0/more/getting_started/unix-variants.html
* $ yum -y install python-devel
* $ tar jxvf boost_1_58_0.tar.bz2
* $ cd boost_1_58_0
* $ ./bootstrap.sh
* $ sudo ./b2 install --build-type=complete --layout=tagged

#### error when running server<br>
######&nbsp; /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.15' not found
* strings /usr/lib64/libstdc++.so.6 | grep GLIBC   //  check the GLIBCXX_3.4.15. It is not in it.
* sudo find / -name "libstdc++.so*" 2>/dev/null  // find the latest libstdc++.so   It should be /usr/local/lib64/libstdc++.so.6.0.18
* sudo rm /usr/lib64/libstdc++.so.6
* sudo cp /usr/local/lib64/libstdc++.so.6.0.18 /usr/lib64/
* sudo ln -s /usr/lib64/libstdc++.so.6.0.18 /usr/lib64/libstdc++.so.6
* sudo vim /etc/ld.so.conf   // add a new line: /usr/lib64
* sudo ldconfig // all ok



### OSX install help
#### gcc 4.8
* install port if do not have one:  https://www.macports.org/install.php
* port search gcc   # show gcc version
* sudo /opt/local/bin/port install gcc48
* sudo /opt/local/bin/port select --set gcc mp-gcc48
* last step,  close old shell, open a new shell.
######https://apple.stackexchange.com/questions/118830/install-gcc-to-mac-via-terminal
######http://www.tuicool.com/articles/reqiY3

#### boost 1.58
* https://stackoverflow.com/questions/104322/how-do-you-install-boost-on-macos


asio_kcp's People

Contributors

libinzhangyuan avatar

Watchers

 avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.