Refactor documents. Delete all backend documents, cc @peng-yongsheng , considering backend refactored, then the documents should rebuild manually.

docs/en/guides/How-to-release.md

@@ -307,7 +307,7 @@ enter your apache password
 3. Public download URLs under `http://www.apache.org/dyn/closer.cgi/incubator/skywalking/x.y.z`.
 4. Public KEYS, sigs and sha512 URLs under `https://www.apache.org/dist/incubator/skywalking/xxxx`
 5. Re-tag on GitHub, tag `x.y.z-RCx` as `x.y.z`, and publish a GitHub release based on this tag.
-5. Send ANNOUNCE mail to `general@incubator.apache.org`.
+5. Send ANNOUNCE mail to `general@incubator.apache.org` and `dev@skywalking.apache.org`.
 ```
 Mail title: [ANNOUNCE] Release Apache SkyWalking (incubating) version x.y.z
 

docs/en/setup/README.md

@@ -3,10 +3,10 @@ Setup based on which kind of probes are you going to use. If you don't understan
 
 ## Language agents in Service 
   - Install agents
-    - [Java agent](service-agent/java-agent/README.md). Introduce how to install java agent to your application, without change any codes.
+    - [Java agent](service-agent/java-agent/README.md). Introduce how to install java agent to your service, without change any codes.
     - [.NET Core agent](https://github.com/OpenSkywalking/skywalking-netcore). See .NET Core agent project document for more details.
     - [Node.js agent](https://github.com/OpenSkywalking/skywalking-nodejs). See Node.js server side agent project document for more details.
-  - [Setup backend](service-agent/backend-setup.md). Set the backend for language agents scenario.
+  - [Setup backend](backend/backend-language-agent-setup.md). Set the backend for language agents scenario.
   
 ## On Service Mesh
   - Istio

docs/en/setup/service-agent/java-agent/Applicaton-toolkit.md

@@ -1,14 +0,0 @@
-# What's sky-walking application toolkit?
-Sky-walking application toolkit is a batch of libraries, provided by skywalking APM. Using them, you have a bridge between your application and skywalking APM agent. 
-
-_**Most important**_, they will not trigger any runtime or performance issues for your application, whether skywalking tracer is active or not. 
-
-# What does bridge mean?
-As you known, skywalking agent run by -javeagent VM parameter. So you definitely don't need to change even a single line of your codes. But in some cases, you want to do interop with tracing/APM system. This is the moment you want to use application toolkit. 
-e.g.
-1. Integrate trace context(traceId) into your log component, e.g. log4j, log4j2 and logback. 
-1. Use CNCF OpenTracing for manually instrumentation. 
-1. Use Skywalking annotation and interactive APIs. 
-
-
-_**Notice**: all toolkits libraries are on bitray.com/jcenter. And make sure their version should be as same as the tracer's version._

docs/en/setup/service-agent/java-agent/Direct-uplink.md

@@ -1,32 +0,0 @@
-# Direct uplink
-## Supported version
-5.0.0-beta +
-
-## What is direct uplink?
-In default, SkyWalking agent uses naming service to find all collector addresses. Then connect to gRPC services.
-
-In **direct uplink**, mean no naming service available or don't work well, set the gRPC addresses in agent directly. 
-The agent uses theses addresses to connect gRPC services.
-
-## Why need this?
-Agent uplink data through the following network
-1. VPCs in Cloud
-1. Internet
-1. Different subnet.
-1. IPs and Ports proxy
-
-## Set the agent config
-1. Remove `collector.servers` config item.
-2. You can find the following settings in `agent.config`
-```
-# Collector agent_gRPC/grpc service addresses.
-# Secondary address setting, only effect when "collector.servers" is empty.
-# By using this, no discovery mechanism provided. The agent only uses these addresses to uplink data.
-# Recommend to use this only when collector cluster IPs are unreachable from agent side. Such as:
-#   1. Agent and collector cluster are in different VPC in Cloud.
-#   2. Agent uplinks data to collector cluster through Internet.
-# collector.direct_servers=www.skywalking.service.io
-```  
-
-3. Set `collector.direct_servers` to a domain name, IP:PORTs, with split by comma(,).
-

docs/en/setup/service-agent/java-agent/Namespace.md

@@ -1,14 +1,12 @@
 # Namespace
-## Supported version
-5.0.0-beta +
 
 ## Background
 SkyWalking is a monitoring tool, which collects metrics from a distributed system. In the real world, a very large distributed system
-includes hundreds of application, thousands of application instance. In that case, more likely, more than one group, 
+includes hundreds of services, thousands of service instances. In that case, most likely, more than one group, 
-even than one company are maintaining and monitoring the distributed system. Each one of them takes charge of different parts,
+even more than one company are maintaining and monitoring the distributed system. Each one of them takes charge of different parts,
-don't or can't share there metrics.
+don't want or shouldn't share there metrics.
 
-Namespace is proposal from this.It is used for tracing and monitoring isolation.
+Namespace is the proposal from this.It is used for tracing and monitoring isolation.
 
 ## Set the namespace
 ### Set agent.namespace in agent config
@@ -21,19 +19,6 @@ The default value of `agent.namespace` is empty.
 
 **Influence**
 The default header key of SkyWalking is `sw3`, more in this [document](../../../protocols/Skywalking-Cross-Process-Propagation-Headers-Protocol-v1.md).
-After `agent.namespace` set, the key changes to `namespace:sw3`.
+After `agent.namespace` set, the key changes to `namespace-sw3`.
 
 The across process propagation chain breaks, when the two sides are using different namespace.
-
-### Set namespace in collector
-```yml
-configuration:
-  default:
-    namespace: xxxxx
-```
-
-**Influences**
-1. If cluster model is active, with zookeeper implementation, The path in zookeeper is changed to include namespace prefix path.
-1. If use Elasticsearch as storage implementation, all type names are changed to include namespace prefix.
-
-

docs/en/setup/service-agent/java-agent/README.md

@@ -1,6 +1,6 @@
 # Setup java agent
-1. Find `agent` folder in SkyWalking 
+1. Find `agent` folder in SkyWalking release package
-2. Add `-javaagent:/path/to/skywalking-agent/skywalking-agent.jar` to VM argument. 
+2. Add `-javaagent:/path/to/skywalking-package/agent/skywalking-agent.jar` to JVM argument. And make sure to add it before the `-jar` argument.
 
 New agent package looks like this：
 ```
@@ -39,7 +39,7 @@ Change the first line of `tomcat/bin/catalina.bat`.
 set "CATALINA_OPTS=-javaagent:/path/to/skywalking-agent/skywalking-agent.jar"
 ```
 - JAR file  
-Add `-javaagent` argument to command line in which you start your app. And make sure to add it before the `-jar` argument. eg:
+Add `-javaagent` argument to command line in which you start your app. eg:
  ```shell
  java -javaagent:/path/to/skywalking-agent/skywalking-agent.jar -jar yourApp.jar
  ```
@@ -56,20 +56,16 @@ Now, we have the following known plugins.
 * [Filter traces through custom services](agent-optional-plugins/trace-ignore-plugin.md)
 
 ## Advanced Features
-* [Override settings through System.properties](en/java-agent/Setting-override.md)
+* Set the settings through system properties for config file override. Read [setting override](Setting-override.md).
-* [Direct uplink and disable naming discovery](en/java-agent/Direct-uplink.md)
+* Use gRPC TLS to link backend. See [open TLS](TLS.md)
-* [Open TLS](en/java-agent/TLS.md)
+* Monitor a big cluster by different SkyWalking services. Use [Namespace](Namespace.md) to isolate the context propagation. 
-* [Namespace Isolation](en/java-agent/Namespace.md)
+* Set client [token](Token-auth.md) if backend open [token authentication](../../backend/backend-token-auth.md).
-* [Token Authentication](en/java-agent/Token-auth.md)
+* Application Toolkit, are a collection of libraries, provided by skywalking APM. Using them, you have a bridge between your application and skywalking APM agent. 
-* Application Toolkit
+    * If you want to use OpenTracing Java APIs, try [SkyWalking OpenTracing compatible tracer](Opentracing.md). More details you could find at http://opentracing.io
-    * [Overview](en/java-agent/Applicaton-toolkit.md)
+    * If you want to print trace context(e.g. traceId) in your logs, choose the log frameworks, [log4j](Application-toolkit-log4j-1.x.md), 
-    * [Use SkyWalking OpenTracing compatible tracer](en/java-agent/Opentracing.md)
+[log4j2](Application-toolkit-log4j-2.x.md), [logback](Application-toolkit-logback-1.x.md)
-    * Integration with log frameworks
+    * If you want to use annotations or SkyWalking native APIs to read context, try [SkyWalking manual APIs](Application-toolkit-trace.md)
-      * [log4j](en/java-agent/Application-toolkit-log4j-1.x.md)
+    * If you want to continue traces across thread manually, use [across thread solution APIs](Application-toolkit-trace-cross-thread.md).
-      * [log4j2](en/java-agent/Application-toolkit-log4j-2.x.md)
-      * [logback](en/java-agent/Application-toolkit-logback-1.x.md)
-    * [Trace by SkyWalking manual APIs](en/java-agent/Application-toolkit-trace.md)
-    * [Trace across threads](en/java-agent/Application-toolkit-trace-cross-thread.md)
 
 # Test
 If you are interested in plugin compatible tests or agent performance, see the following reports.

docs/en/setup/service-agent/java-agent/TLS.md

@@ -7,18 +7,9 @@ at the same time, the SkyWalking backend is in another region (VPC).
 > 
 > Because of that, security requirement is very obvious.
 
-## Requirement
-Enable **direct uplink**, by following this [document](Direct-uplink.md).
-
-Because of uplink through internet, with security concern, the naming mechanism didn't fit. 
-So we didn't support TLS in naming service of HTTP service.
-
-## Supported version
-5.0.0-beta +
-
 ## Authentication Mode
 Only support **no mutual auth**.
-- Use this [script](../../../tools/TLS/tls_key_generate.sh) if you are not familiar with how to generate key files.
+- Use this [script](../../../../../tools/TLS/tls_key_generate.sh) if you are not familiar with how to generate key files.
 - Find `ca.crt`, and use it at client side
 - Find `server.crt` and `server.pem`. Use them at server side.
 
@@ -29,18 +20,4 @@ Only support **no mutual auth**.
 
 Agent open TLS automatically after the `/ca/ca.crt` file detected.
 
-### Collector config
+o make sure can't access other ports out of region (VPC), such as firewall, proxy.
-Module `agent_gRPC/gRPC` supports TLS. And only this module for now.
-
-- Uncomment the `ssl_cert_chain_file` and `ssl_private_key_file` settings in `application.yml`
-- `ssl_cert_chain_file` value is the absolute path of `server.crt`
-- `ssl_private_key_file` value is the absolute path of `server.pem`
-
-## Avoid port share
-In most cases, we recommend sharing port for all gRPC services in `agent_gRPC/gRPC` and `remote/gRPC` modules.
-But don't do this when you open TLS in `agent_gRPC/gRPC`, the obvious reason is you can't listen a port with and without TLS.
-
-The solution is, change the `remote/gRPC/port`.
-
-## How about other listening ports
-Please use other security ways to make sure can't access other ports out of region (VPC), such as firewall, proxy.

docs/en/setup/service-agent/java-agent/Token-auth.md

@@ -1,31 +1,16 @@
 # Token Authentication
-## Supported version
-5.0.0-beta +
-
-## How need token authentication after we have TLS?
-TLS is about transport security, which makes sure the network can be trusted. 
-The token authentication is about monitoring application data **can be trusted**.
 
 ## Token 
 In current version, Token is considered as a simple string.
 
 ### Set Token
-1. Set token in agent.config file
+Set token in agent.config file
 ```properties
 # Authentication active is based on backend setting, see application.yml for more details.
 agent.authentication = xxxx
 ```
 
-2. Set token in application.yml file
+Meanwhile, open the [backend token authentication](../../backend/backend-token-auth.md).
-```yaml
-agent_gRPC:
-  gRPC:
-    host: localhost
-    port: 11800
-
-    #Set your own token to active auth
-    authentication: xxxxxx
-```
 
 ## Authentication fails
 The Collector verifies every request from agent, allowed only the token match.

docs/en/setup/service-agent/java-agent/Use-ShardingJDBC-as-storage-implementor.md

@@ -1,21 +0,0 @@
-# Supported Database Sharding Storage
-Beside the default Elasticsearch storage, the user can use shardingJDBC and MySQL as storage implementor.
-Note: it only supported MYSQL database sharding, and due to the license restrictions, the users need to import MYSQL Driver manually.
-
-## Requirement
-- Manually import MySQL Driver package mysql-connector-java-5.1.36.jar to collector libs directory.
-- In config/application.yml, delete the ElasticSearch configuration and add the following shardingjdbc configuration.
-```
-  storage:
-    shardingjdbc:
-      driverClass: com.mysql.jdbc.Driver
-      # JDBC Datasource connections for ShardingJDBC, multiple should be separated by comma
-      url: jdbc:mysql://ip1:port1/skywalking,jdbc:mysql://ip2:port2/skywalking
-      # Usernames, which match the sequence of Datasource URLs
-      userName: admin,admin
-      # Passwords, which match the sequence of Datasource URLs
-      password: 123456,123456
-```
-
-## See also
-- ShardingJDBC now named as ShardingSphere since its 3.x, see [here](http://shardingsphere.io)