From 86399a8927b1efff36d1a56eeb411c4ed0948b9b Mon Sep 17 00:00:00 2001
From: KernelMaker <songzhao.asm@icloud.com>
Date: Fri, 22 Apr 2016 15:06:14 +0800
Subject: [PATCH] update README.md

---
 README.md    | 56 +++++++++++++++++++++++++++++++++++++---------------
 README_CN.md | 25 -----------------------
 README_EN.md |  3 +++
 3 files changed, 43 insertions(+), 41 deletions(-)
 delete mode 100644 README_CN.md
 create mode 100644 README_EN.md

diff --git a/README.md b/README.md
index 481f7c66..dfa68e14 100644
--- a/README.md
+++ b/README.md
@@ -1,22 +1,46 @@
 # Pika
-## Introduction [中文] (https://github.com/baotiao/pika/blob/pika2.0/README_CN.md)
-[![Join the chat at https://gitter.im/baotiao/pika](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/baotiao/pika?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)
-Pika is a database that has the same feature as redis. And Pika is compability
-with redis client. 
+## 简介 [English](https://github.com/baotiao/pika/tree/pika2.0)
+Pika是一个可持久化的大容量redis存储服务，兼容string、hash、list、zset、set的绝大接口([兼容详情](https://github.com/baotiao/pika/wiki/pika支持的redis接口及兼容情况))，解决redis由于存储数据量巨大而导致内存不够用的容量瓶颈，并且可以像redis一样，通过slaveof命令进行主从备份，支持全同步和部分同步
 
-The reason why we develop Pika is to solve redis's memory limitation
+## 特点
+* 容量大，支持百G数据量的存储
+* 兼容redis，不用修改代码即可平滑从redis迁移到pika
+* 支持主从(slaveof)
+* 完善的[运维](https://github.com/baotiao/pika/wiki/pika的一些管理命令方式说明)命令
 
-Pika used Nemo as storage engine, Nemo is developed base on rocksdb.
+## 编译安装
+```
+1. git submodule init && git submodule update
+2. make __REL=1 (编译依赖的某些库如snappy，bz2请自行提前安装）
+3. 将./lib/_VERSION_/lib目录移动到Makefile中自定义的rpath路径，pika启动时会从rpath加载相关so
+```
 
-Nemo support multi data structure by using rocksdb's kv port. Such as list,
-hash, zset, set
+## 使用
+```
+./output/bin/pika -c ./conf/pika.conf
+```
 
-Because rocksdb is using disk, so Nemo don't have capacity limitation.
+## 性能
+```
+服务端配置：
+	处理器：24核 Intel(R) Xeon(R) CPU E5-2630 v2 @ 2.60GHz
+	内存：165157944 kB
+	操作系统：CentOS release 6.2 (Final)
+	网卡：Intel Corporation I350 Gigabit Network Connection
+客户端配置：
+	同服务端
+	
+测试结果：
+	pika配置18个worker，用40个客户端；
+	1. 写性能测试：
+		方法：客户端依次执行set、hset、lpush、zadd、sadd接口写入数据，每个数据结构10000个key；
+		结果：qps 110000
+	2. 读性能测试：
+		方法：客户端一次执行get、hget、lindex、zscore、smembers，每个数据结构5000000个key；
+		结果：qps 170000
+```
+## 文档
+1. [Wiki] (https://github.com/baotiao/pika/wiki)
 
-The protocol used to contact with Pika is redis protocol, so you don't need to
-change your code to migrate to Pika
-
-### Thread Model
-
-Pika used multi threads model. The main thread accept the request and then send
-the request to the work threads.
+## 联系方式
+songzhao@360.cn
diff --git a/README_CN.md b/README_CN.md
deleted file mode 100644
index cb2207fe..00000000
--- a/README_CN.md
+++ /dev/null
@@ -1,25 +0,0 @@
-# Pika
-## 简介 [English](https://github.com/baotiao/pika/tree/pika2.0)
-Pika是一个可持久化的大容量redis存储服务，兼容string、hash、list、zset、set的绝大接口，解决redis由于存储数据量巨大而导致内存不够用的容量瓶颈，并且可以像redis一样，通过slaveof命令进行主从备份，支持全同步和部分同步
-
-## 特点
-* 容量大，支持百G数据量的存储
-* 兼容redis，不用修改代码即可平滑从redis迁移到pika
-* 支持主从
-
-## 编译安装
-```
-git submodule update
-make __REL=1
-```
-
-## 使用
-```
-./output/bin/pika -c ./conf/pika.conf
-```
-
-## 性能
-
-## 文档
-
-## 联系方式
\ No newline at end of file
diff --git a/README_EN.md b/README_EN.md
new file mode 100644
index 00000000..2040e482
--- /dev/null
+++ b/README_EN.md
@@ -0,0 +1,3 @@
+# Pika
+## Introduction [中文] (https://github.com/baotiao/pika/blob/pika2.0/README_EN.md)
+Coming Soon...
-- 
GitLab