diff --git a/docs-2.0-en/3.ngql-guide/5.operators/4.pipe.md b/docs-2.0-en/3.ngql-guide/5.operators/4.pipe.md
index 8bd8c691714..edd8383474a 100644
--- a/docs-2.0-en/3.ngql-guide/5.operators/4.pipe.md
+++ b/docs-2.0-en/3.ngql-guide/5.operators/4.pipe.md
@@ -39,9 +39,7 @@ In NebulaGraph, pipes will affect the performance. Take `A | B` as an example, t
 
 1. Pipe operators operate synchronously. That is, the data can enter the pipe clause as a whole after the execution of clause `A` before the pipe operator is completed.
 
-2. Pipe operators need to be serialized and deserialized, which is executed in a single thread.
-
-3. If `A` sends a large amount of data to `|`, the entire query request may be very slow. You can try to split this statement.
+2. If `A` sends a large amount of data to `|`, the entire query request may be very slow. You can try to split this statement.
 
   1. Send `A` from the application,
 
diff --git a/docs-2.0-zh/3.ngql-guide/5.operators/4.pipe.md b/docs-2.0-zh/3.ngql-guide/5.operators/4.pipe.md
index bbeec9a32f6..546605596c7 100644
--- a/docs-2.0-zh/3.ngql-guide/5.operators/4.pipe.md
+++ b/docs-2.0-zh/3.ngql-guide/5.operators/4.pipe.md
@@ -39,9 +39,7 @@ nebula> GO FROM "player100" OVER follow \
 
 1. 管道是同步操作。也即需要管道之前的子句`A`执行完毕后，数据才能整体进入管道子句。
 
-2. 管道本身是需要序列化和反序列化的，这个是单线程执行的。
-
-3. 如果`A`发大量数据给 `|`，整个查询请求的总体时延可能会非常大。此时可以尝试拆分这个语句：
+2. 如果`A`发大量数据给 `|`，整个查询请求的总体时延可能会非常大。此时可以尝试拆分这个语句：
 
   1. 应用程序发送`A`，