diff --git a/core/docs/en/ordered-work-steal.md b/core/docs/en/ordered-work-steal.md
new file mode 100644
index 00000000..95b973e7
--- /dev/null
+++ b/core/docs/en/ordered-work-steal.md
@@ -0,0 +1,59 @@
+---
+title: Ordered Work Steal Overview
+date: 2025-01-17 08:34:00
+author: loongs-zhang
+---
+
+# Ordered Work Steal Overview
+
+## Why ordered work steal?
+
+In the real world, there are always threads that complete their own tasks first, while other threads have tasks to be
+processed. Then, a spectacular scene emerged where one core was in difficulty and other cores were watching.
+
+<div style="text-align: center;">
+    <img src="../../../docs/img/watching.png" width="50%">
+</div>
+
+Obviously, we don't want this to happen. For idle threads, instead of letting them watch other threads working, it's
+better to let them help other threads work. In addition, we hope to pop tasks based on priority, the higher the
+priority, the earlier it will be popped up.
+
+## What is ordered work steal queue?
+
+An ordered work steal queue consists of a global queue and multiple local queues, the global queue is unbounded, while
+the local queue has a bounded SkipList with collections. To ensure high performance, the number of local queues is
+usually equal to the number of threads.
+
+## How `push` works
+
+```mermaid
+flowchart TD
+    Cond{Is the local queue full?}
+    PS[Push to the local queue]
+    PTG[Push half of the low priority tasks from the local queue to the global queue]
+    PG[Push to the global queue]
+    push --> Cond
+    Cond -- No --> PS
+    Cond -- Yes --> PTG --- PG
+    
+```
+
+## How `pop` works
+
+```mermaid
+flowchart TD
+    Cond1{Are there any tasks in the local queue?}
+    Cond2{Are there any tasks in other local queues?}
+    Cond3{Are there any tasks in the global queue?}
+    PS[Pop the task with the highest priority]
+    ST[Steal tasks with high priority from other local queues]
+    PF[Task not found]
+    pop --> Cond1
+    Cond1 -- Yes --> PS
+    Cond1 -- No --> Cond2
+    Cond2 -- Yes --> ST --> PS
+    Cond2 -- No --> Cond3
+    Cond3 -- Yes --> PS
+    Cond3 -- No --> PF
+```
diff --git a/core/docs/en/work-steal.md b/core/docs/en/work-steal.md
new file mode 100644
index 00000000..96281690
--- /dev/null
+++ b/core/docs/en/work-steal.md
@@ -0,0 +1,58 @@
+---
+title: Work Steal Overview
+date: 2025-01-17 08:34:00
+author: loongs-zhang
+---
+
+# Work Steal Overview
+
+## Why work steal?
+
+In the real world, there are always threads that complete their own tasks first, while other threads have tasks to be
+processed. Then, a spectacular scene emerged where one core was in difficulty and other cores were watching.
+
+<div style="text-align: center;">
+    <img src="../../../docs/img/watching.png" width="50%">
+</div>
+
+Obviously, we don't want this to happen. For idle threads, instead of letting them watch other threads working, it's
+better to let them help other threads work.
+
+## What is work steal queue?
+
+A work steal queue consists of a global queue and multiple local queues, the global queue is unbounded, while the local
+queue has a bounded RingBuffer. To ensure high performance, the number of local queues is usually equal to the number of
+threads.
+
+## How `push` works
+
+```mermaid
+flowchart TD
+    Cond{Is the local queue full?}
+    PS[Push to the local queue]
+    PTG[Push half of the tasks from the local queue to the global queue]
+    PG[Push to the global queue]
+    push --> Cond
+    Cond -- No --> PS
+    Cond -- Yes --> PTG --- PG
+    
+```
+
+## How `pop` works
+
+```mermaid
+flowchart TD
+    Cond1{Are there any tasks in the local queue?}
+    Cond2{Are there any tasks in other local queues?}
+    Cond3{Are there any tasks in the global queue?}
+    PS[Pop a stack]
+    ST[Steal tasks from other local queues]
+    PF[Task not found]
+    pop --> Cond1
+    Cond1 -- Yes --> PS
+    Cond1 -- No --> Cond2
+    Cond2 -- Yes --> ST --> PS
+    Cond2 -- No --> Cond3
+    Cond3 -- Yes --> PS
+    Cond3 -- No --> PF
+```
diff --git a/core/src/common/mod.rs b/core/src/common/mod.rs
index f92e005f..c37551fd 100644
--- a/core/src/common/mod.rs
+++ b/core/src/common/mod.rs
@@ -51,6 +51,7 @@ pub mod beans;
 /// assert_eq!(queue.pop(), None);
 /// ```
 ///
+#[doc = include_str!("../../docs/en/work-steal.md")]
 pub mod work_steal;
 
 /// Suppose a thread in a work-stealing scheduler is idle and looking for the next task to run. To
@@ -87,6 +88,7 @@ pub mod work_steal;
 /// assert_eq!(queue.pop(), None);
 /// ```
 ///
+#[doc = include_str!("../../docs/en/ordered-work-steal.md")]
 pub mod ordered_work_steal;
 
 #[cfg(target_os = "linux")]
diff --git a/docs/img/watching.png b/docs/img/watching.png
new file mode 100644
index 00000000..f5f44caf
Binary files /dev/null and b/docs/img/watching.png differ