diff --git a/docs-2.0/nebula-exchange/about-exchange/ex-ug-what-is-exchange.md b/docs-2.0/nebula-exchange/about-exchange/ex-ug-what-is-exchange.md
index 52752caaccd..64589ff7ed8 100644
--- a/docs-2.0/nebula-exchange/about-exchange/ex-ug-what-is-exchange.md
+++ b/docs-2.0/nebula-exchange/about-exchange/ex-ug-what-is-exchange.md
@@ -52,6 +52,8 @@ Exchange has the following advantages:
 
 ## Version compatibility
 
+Exchange supports Spark versions 2.2.x, 2.4.x, and 3.x.x, which are named `nebula-exchange_spark_2.2`, `nebula-exchange_spark_2.4`, and `nebula-exchange_spark_3.0` for different Spark versions.
+
 The correspondence between the NebulaGraph Exchange version (the JAR version), the NebulaGraph core version and the Spark version is as follows.
 
 | Exchange version | NebulaGraph version | Spark version |
diff --git a/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-jdbc.md b/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-jdbc.md
index 9ce1c536b79..5dab980dda6 100644
--- a/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-jdbc.md
+++ b/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-jdbc.md
@@ -80,6 +80,10 @@ Before importing data, you need to confirm the following information:
 
 - The Hadoop service has been installed and started.
 
+## Precautions
+
+nebula-exchange_spark_2.2 supports only single table queries, not multi-table queries.
+
 ## Steps
 
 ### Step 1: Create the Schema in NebulaGraph
@@ -194,11 +198,18 @@ After Exchange is compiled, copy the conf file `target/classes/application.conf`
       driver:"com.mysql.cj.jdbc.Driver"
 
       # Database user name and password
-      user:root
+      user:"root"
       password:"12345"
 
-      table:player
-      sentence:"select playerid, age, name from player order by playerid"
+      # Scanning a single table to read data.
+      # nebula-exchange_spark_2.2 must configure this parameter, and can additionally configure sentence.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as sentence.
+      table:"basketball.player"
+
+      # Use query statement to read data.
+      # nebula-exchange_spark_2.2 can configure this parameter. Multi-table queries are not supported. Only the table name needs to be written after from. The form `db.table` is not supported.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as table. Multi-table queries are supported.
+      # sentence:"select playerid, age, name from player, team order by playerid"
 
       # (optional)Multiple connections read parameters. See https://spark.apache.org/docs/latest/sql-data-sources-jdbc.html
       partitionColumn:playerid    # optional. Must be a numeric, date, or timestamp column from the table in question.
@@ -282,8 +293,17 @@ After Exchange is compiled, copy the conf file `target/classes/application.conf`
       driver:"com.mysql.cj.jdbc.Driver"
       user:root
       password:"12345"
-      table:follow
-      sentence:"select src_player,dst_player,degree from follow order by src_player"
+
+      # Scanning a single table to read data.
+      # nebula-exchange_spark_2.2 must configure this parameter, and can additionally configure sentence.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as sentence.
+      table:"basketball.follow"
+
+      # Use query statement to read data.
+      # nebula-exchange_spark_2.2 can configure this parameter. Multi-table queries are not supported. Only the table name needs to be written after from. The form `db.table` is not supported.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as table. Multi-table queries are supported.
+      # sentence:"select src_player,dst_player,degree from follow order by src_player"
+
       partitionColumn:src_player    
       lowerBound:1                
       upperBound:5                
diff --git a/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-kafka.md b/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-kafka.md
index c5e20c5e954..9be12c4cf7e 100644
--- a/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-kafka.md
+++ b/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-kafka.md
@@ -46,7 +46,9 @@ Before importing data, you need to confirm the following information:
 
 ## Precautions
 
-Only client mode is supported when importing Kafka data, i.e. the value of parameters `tags.type.sink` and `edges.type.sink` is `client`.
+- Only client mode is supported when importing Kafka data, i.e. the value of parameters `tags.type.sink` and `edges.type.sink` is `client`.
+
+- When importing Kafka data, do not use Exchange version 3.4.0, which adds caching of imported data and does not support streaming data import. Use Exchange versions 3.0.0, 3.3.0, or 3.5.0.
 
 ## Steps
 
diff --git a/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-mysql.md b/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-mysql.md
index 76e075378dd..40d45a6e620 100644
--- a/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-mysql.md
+++ b/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-mysql.md
@@ -82,6 +82,10 @@ Before importing data, you need to confirm the following information:
 
 - The Hadoop service has been installed and started.
 
+## Precautions
+
+nebula-exchange_spark_2.2 supports only single table queries, not multi-table queries.
+
 ## Steps
 
 ### Step 1: Create the Schema in NebulaGraph
@@ -191,11 +195,19 @@ After Exchange is compiled, copy the conf file `target/classes/application.conf`
 
       host:192.168.*.*
       port:3306
-      database:"basketball"
-      table:"player"
       user:"test"
       password:"123456"
-      sentence:"select playerid, age, name from player order by playerid;"
+      database:"basketball"
+
+      # Scanning a single table to read data.
+      # nebula-exchange_spark_2.2 must configure this parameter. Sentence is not supported.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as sentence.
+      table:"basketball.player"
+
+      # Use query statement to read data.
+      # This parameter is not supported by nebula-exchange_spark_2.2.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as table. Multi-table queries are supported.
+      # sentence: "select * from people, player, team"
 
       # Specify the column names in the player table in fields, and their corresponding values are specified as properties in the NebulaGraph.
       # The sequence of fields and nebula.fields must correspond to each other.
@@ -264,11 +276,19 @@ After Exchange is compiled, copy the conf file `target/classes/application.conf`
 
       host:192.168.*.*
       port:3306
-      database:"basketball"
-      table:"follow"
       user:"test"
       password:"123456"
-      sentence:"select src_player,dst_player,degree from follow order by src_player;"
+      database:"basketball"
+
+      # Scanning a single table to read data.
+      # nebula-exchange_spark_2.2 must configure this parameter. Sentence is not supported.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as sentence.
+      table:"basketball.follow"
+
+      # Use query statement to read data.
+      # This parameter is not supported by nebula-exchange_spark_2.2.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as table. Multi-table queries are supported.
+      # sentence: "select * from follow, serve"
 
       # Specify the column names in the follow table in fields, and their corresponding values are specified as properties in the NebulaGraph.
       # The sequence of fields and nebula.fields must correspond to each other.
diff --git a/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-oracle.md b/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-oracle.md
index b2a8b98caa0..d8440d7697d 100644
--- a/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-oracle.md
+++ b/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-oracle.md
@@ -82,6 +82,10 @@ Before importing data, you need to confirm the following information:
 
 - The Hadoop service has been installed and started.
 
+## Precautions
+
+nebula-exchange_spark_2.2 supports only single table queries, not multi-table queries.
+
 ## Steps
 
 ### Step 1: Create the Schema in NebulaGraph
@@ -189,12 +193,20 @@ After Exchange is compiled, copy the conf file `target/classes/application.conf`
         sink: client
       }
 
-      url:"jdbc:oracle:thin:@host:1521:db"
+      url:"jdbc:oracle:thin:@host:1521:basketball"
       driver: "oracle.jdbc.driver.OracleDriver"
       user: "root"
       password: "123456"
-      table: "basketball.player"
-      sentence: "select playerid, name, age from player"
+
+      # Scanning a single table to read data.
+      # nebula-exchange_spark_2.2 must configure this parameter. Sentence is not supported.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as sentence.
+      table:"basketball.player"
+
+      # Use query statement to read data.
+      # This parameter is not supported by nebula-exchange_spark_2.2.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as table. Multi-table queries are supported.
+      # sentence: "select * from people, player, team"
 
       # Specify the column names in the player table in fields, and their corresponding values are specified as properties in the NebulaGraph.
       # The sequence of fields and nebula.fields must correspond to each other.
@@ -226,7 +238,7 @@ After Exchange is compiled, copy the conf file `target/classes/application.conf`
         sink: client
       }
 
-      url:"jdbc:oracle:thin:@host:1521:db"
+      url:"jdbc:oracle:thin:@host:1521:basketball"
       driver: "oracle.jdbc.driver.OracleDriver"
       user: "root"
       password: "123456"
@@ -260,12 +272,20 @@ After Exchange is compiled, copy the conf file `target/classes/application.conf`
         sink: client
       }
 
-      url:"jdbc:oracle:thin:@host:1521:db"
+      url:"jdbc:oracle:thin:@host:1521:basketball"
       driver: "oracle.jdbc.driver.OracleDriver"
       user: "root"
       password: "123456"
-      table: "basketball.follow"
-      sentence: "select src_player, dst_player, degree from follow"
+
+      # Scanning a single table to read data.
+      # nebula-exchange_spark_2.2 must configure this parameter. Sentence is not supported.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as sentence.
+      table:"basketball.follow"
+
+      # Use query statement to read data.
+      # This parameter is not supported by nebula-exchange_spark_2.2.
+      # nebula-exchange_spark_2.4 and nebula-exchange_spark_3.0 can configure this parameter, but not at the same time as table. Multi-table queries are supported.
+      # sentence: "select * from follow, serve"
 
       # Specify the column names in the follow table in fields, and their corresponding values are specified as properties in the NebulaGraph.
       # The sequence of fields and nebula.fields must correspond to each other.
@@ -311,7 +331,7 @@ After Exchange is compiled, copy the conf file `target/classes/application.conf`
         sink: client
       }
 
-      url:"jdbc:oracle:thin:@host:1521:db"
+      url:"jdbc:oracle:thin:@host:1521:basketball"
       driver: "oracle.jdbc.driver.OracleDriver"
       user: "root"
       password: "123456"
diff --git a/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-pulsar.md b/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-pulsar.md
index 2699747fda3..af27124bdd1 100644
--- a/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-pulsar.md
+++ b/docs-2.0/nebula-exchange/use-exchange/ex-ug-import-from-pulsar.md
@@ -34,7 +34,9 @@ Before importing data, you need to confirm the following information:
 
 ## Precautions
 
-Only client mode is supported when importing Pulsar data, i.e. the value of parameters `tags.type.sink` and `edges.type.sink` is `client`.
+- Only client mode is supported when importing Pulsar data, i.e. the value of parameters `tags.type.sink` and `edges.type.sink` is `client`.
+
+- When importing Pulsar data, do not use Exchange version 3.4.0, which adds caching of imported data and does not support streaming data import. Use Exchange versions 3.0.0, 3.3.0, or 3.5.0.
 
 ## Steps