Update Chinese Doc duplication and translate it to English

[ninsmiracle]

No description provided.

[acelyc111]

Hi @ninsmiracle , many thanks for the contribution!

AFAIK, the Chinese docs are a few outdated, for example:
1.

目前在跨机房网络健康的环境下，数据延时大概在 10s 左右，即 A 机房的写数据大概在 10s 后会写入 B 机房。

The 10s can be configured now (by apache/incubator-pegasus#1639), you can add describption like:

Since Pegasus 2.6, it can by configured by xxx

2. Because FDS is not supported since 2.6, so it would be better to use HDFS as an example to backup data.
3. Because the metric names have been updated since 2.6, it would be better to distinguish the metric names. Or we can do the update in another patch.
4. "完整监控项/HTTP 接口列表" can be removed, because it's not "完整的" and it will cost much time to maintenance the doc when code update.

[acelyc111]

It's recommand to update the Chinese docs at first, then translate it to English to reduce duplicate work.

[ninsmiracle]

目前在跨机房网络健康的环境下，数据延时大概在 10s 左右，即 A 机房的写数据大概在 10s 后会写入 B 机房。

In my opinion,duplication_sync_period_second is the parameter who control the interval time that the duplication info between meta server and replica server. Intead of control the interval of data dup.
Cause we made some latency tests between master duplication cluster and backup duplication cluster.We write some k-v data to master cluster and each of them less than 1KB,Our results showed that the backup duplication cluster was able to read the data in just about 100ms.

[acelyc111]

目前在跨机房网络健康的环境下，数据延时大概在 10s 左右，即 A 机房的写数据大概在 10s 后会写入 B 机房。

In my opinion,duplication_sync_period_second is the parameter who control the interval time that the duplication info between meta server and replica server. Intead of control the interval of data dup. Cause we made some latency tests between master duplication cluster and backup duplication cluster.We write some k-v data to master cluster and each of them less than 1KB,Our results showed that the backup duplication cluster was able to read the data in just about 100ms.

Thanks for the information! Is it the fixed time interval in https://github.com/apache/incubator-pegasus/blob/master/src/replica/duplication/duplication_pipeline.cpp#L63 ?

The docs is needed to be updated anyway.

[acelyc111]

apache/incubator-pegasus#1745

[ninsmiracle]

THX for the comment and sorrry about the untimely reply. Cause we have too many online failure recently...