diff --git a/tutorials/training/source_en/advanced_use/performance_profiling_ascend.md b/tutorials/training/source_en/advanced_use/performance_profiling_ascend.md
index ea4f1799b71ffed0cd57339969adf31be8c4a9b0..21eab0daa2eb94b230c45d71784a578581c033d2 100644
--- a/tutorials/training/source_en/advanced_use/performance_profiling_ascend.md
+++ b/tutorials/training/source_en/advanced_use/performance_profiling_ascend.md
@@ -261,4 +261,5 @@ Figure 9：Memory Statistics
 ## Notices
 
 - Currently running in PyNative mode is not supported.
-- Currently the training and inference process does not support performance debugging, only individual training or inference is supported.
\ No newline at end of file
+- Currently the training and inference process does not support performance debugging, only individual training or inference is supported.
+- Currently training Heterogeneous network in "dataset_sink_mode=True" mode is not supported, performance debugging is not supported in this scene.
\ No newline at end of file
diff --git a/tutorials/training/source_en/advanced_use/performance_profiling_ascend_of_cluster.md b/tutorials/training/source_en/advanced_use/performance_profiling_ascend_of_cluster.md
index d328233b9a81b3db947412619632651670ebd370..b2167577a77f41561905bed3bb52c947ad25f335 100644
--- a/tutorials/training/source_en/advanced_use/performance_profiling_ascend_of_cluster.md
+++ b/tutorials/training/source_en/advanced_use/performance_profiling_ascend_of_cluster.md
@@ -237,3 +237,4 @@ Figure 2 shows the performance information of a single device in the cluster. Pl
 
 - Currently running in PyNative mode is not supported.
 - Currently the training and inference process does not support performance debugging, only individual training or inference is supported.
+- Currently training Heterogeneous network in "dataset_sink_mode=True" mode is not supported, performance debugging is not supported in this scene.
diff --git a/tutorials/training/source_en/advanced_use/performance_profiling_gpu.md b/tutorials/training/source_en/advanced_use/performance_profiling_gpu.md
index 7a290a05859dbbcc7bf00d3914279c2be29a68e0..80ab340acab67ad2a45aa71da2938f2f651d5072 100644
--- a/tutorials/training/source_en/advanced_use/performance_profiling_gpu.md
+++ b/tutorials/training/source_en/advanced_use/performance_profiling_gpu.md
@@ -175,4 +175,5 @@ The usage is described as follows:
 ## Notices
 
 - Currently running in PyNative mode is not supported.
-- Currently the training and inference process does not support performance debugging, only individual training or inference is supported.
\ No newline at end of file
+- Currently the training and inference process does not support performance debugging, only individual training or inference is supported.
+- Currently training Heterogeneous network in "dataset_sink_mode=True" mode is not supported, performance debugging is not supported in this scene.
\ No newline at end of file
diff --git a/tutorials/training/source_zh_cn/advanced_use/performance_profiling_ascend.md b/tutorials/training/source_zh_cn/advanced_use/performance_profiling_ascend.md
index a1fce40f01b2fddf356f6521dacc211584935c04..5b2fa86f4bb2f1cbe9ab73ee0562bb3c5a287577 100644
--- a/tutorials/training/source_zh_cn/advanced_use/performance_profiling_ascend.md
+++ b/tutorials/training/source_zh_cn/advanced_use/performance_profiling_ascend.md
@@ -263,4 +263,5 @@ CPU利用率常用场景:
 ## 注意事项
 
 - PyNative模式下暂不支持性能调试。
-- 训练加推理过程暂不支持性能调试，目前支持单独训练或推理的性能调试。
\ No newline at end of file
+- 训练加推理过程暂不支持性能调试，目前支持单独训练或推理的性能调试。
+- 异构网络暂不支持数据下沉(dataset_sink_mode=True)模式训练，此场景暂不支持性能调试。
\ No newline at end of file
diff --git a/tutorials/training/source_zh_cn/advanced_use/performance_profiling_ascend_of_cluster.md b/tutorials/training/source_zh_cn/advanced_use/performance_profiling_ascend_of_cluster.md
index 0e811755d72c00f94ac02f79b839bccb5b877e82..1a83f98f4b3bdd8739176cff4d68dfcfb7cfba34 100644
--- a/tutorials/training/source_zh_cn/advanced_use/performance_profiling_ascend_of_cluster.md
+++ b/tutorials/training/source_zh_cn/advanced_use/performance_profiling_ascend_of_cluster.md
@@ -238,4 +238,5 @@ done
 ## 注意事项
 
 - PyNative模式下暂不支持性能调试。
-- 训练加推理过程暂不支持性能调试，目前支持单独训练或推理的性能调试。
\ No newline at end of file
+- 训练加推理过程暂不支持性能调试，目前支持单独训练或推理的性能调试。
+- 异构网络暂不支持数据下沉(dataset_sink_mode=True)模式训练，此场景暂不支持性能调试。
\ No newline at end of file
diff --git a/tutorials/training/source_zh_cn/advanced_use/performance_profiling_gpu.md b/tutorials/training/source_zh_cn/advanced_use/performance_profiling_gpu.md
index 34d360418346abb3c8a94994393da798242aebce..8faed0453b7f895b68e6bc4e920348eefc69a20d 100644
--- a/tutorials/training/source_zh_cn/advanced_use/performance_profiling_gpu.md
+++ b/tutorials/training/source_zh_cn/advanced_use/performance_profiling_gpu.md
@@ -171,4 +171,5 @@ GPU场景下，CPU利用率分析的使用方法和Ascend场景相同。
 ## 注意事项
 
 - PyNative模式下暂不支持性能调试。
-- 训练加推理过程暂不支持性能调试，目前支持单独训练或推理的性能调试。
\ No newline at end of file
+- 训练加推理过程暂不支持性能调试，目前支持单独训练或推理的性能调试。
+- 异构网络暂不支持数据下沉(dataset_sink_mode=True)模式训练，此场景暂不支持性能调试。
\ No newline at end of file