wu-sheng commented on code in PR #12671:
URL: https://github.com/apache/skywalking/pull/12671#discussion_r1824549994


##########
docs/en/setup/backend/backend-async-profiler.md:
##########
@@ -0,0 +1,85 @@
+`# Async Profiler
+
+Async Profiler is bound within the auto-instrument agent and corresponds to 
[In-Process 
Profiling](../../concepts-and-designs/profiling.md#in-process-profiling).
+
+It is passed to the proxy in the form of a task, allowing it to be enabled or 
disabled dynamically.
+When service encounters performance issues (cpu usage, memory allocation, 
locks), async-profiler task can be created.
+When the proxy receives a task, it enables Async Profiler for sampling.
+After sampling is completed, a flame graph will be generated for performance 
analysis to determine the specific business code line that caused the 
performance problem.
+
+## Active in the OAP
+OAP and the agent use a brand-new protocol to exchange Async Profiler data, so 
it is necessary to start OAP with the following configuration:
+
+```yaml
+receiver-async-profiler:
+  selector: ${SW_RECEIVER_ASYNC_PROFILER:default}
+  default:
+```
+
+## Async Profiler Task with Analysis
+
+To use the Async Profiler feature, please follow these steps:
+
+1. **Create Async Profiler task**: Use the UI or CLI tool to create a task.
+2. **Wait agent collect data and upload**: Wait for Async Profiler to collect 
JFR data and report
+3. **Query task progress**: Query the progress of tasks, including analyzing 
successful and failed instances and task logs
+4. **Analyze the data**: Analyze the JFR data to determine where performance 
bottlenecks exist in the service.
+
+### Create Async Profiler task
+
+Create an Async Profiler task to notify some java-agent instances in the 
execution service to start Async Profiler for data collection.
+
+When creating a task, the following configuration fields are required:
+
+1. **serviceId**: Define the service to execute the task.
+2. **serviceInstanceIds**: Define which instances need to execute tasks.
+3. **duration**: Define the duration of this task (second).
+4. **events**: Define which event types this task needs to collect.
+5. **execArgs**: Other Async Profiler execution options, e.g. alloc=2k,lock=2s.
+
+When the Agent receives a Async Profiler task from OAP, it automatically 
generates a log to notify that the task has been acknowledged. The log contains 
the following field information:
+
+1. **Instance**: The name of the instance where the Agent is located.
+2. **Type**: Supports "NOTIFIED" and "EXECUTION_FINISHED" and 
"JFR_UPLOAD_FILE_TOO_LARGE_ERROR", "EXECUTION_TASK_ERROR", with the current log 
displaying "NOTIFIED".
+3. **Time**: The time when the Agent received the task.
+
+### Wait agent collect data and upload

Review Comment:
   ```suggestion
   ### Wait the agent to collect data and upload
   ```



##########
docs/en/setup/backend/backend-async-profiler.md:
##########
@@ -0,0 +1,85 @@
+`# Async Profiler
+
+Async Profiler is bound within the auto-instrument agent and corresponds to 
[In-Process 
Profiling](../../concepts-and-designs/profiling.md#in-process-profiling).
+
+It is passed to the proxy in the form of a task, allowing it to be enabled or 
disabled dynamically.
+When service encounters performance issues (cpu usage, memory allocation, 
locks), async-profiler task can be created.
+When the proxy receives a task, it enables Async Profiler for sampling.
+After sampling is completed, a flame graph will be generated for performance 
analysis to determine the specific business code line that caused the 
performance problem.
+
+## Active in the OAP
+OAP and the agent use a brand-new protocol to exchange Async Profiler data, so 
it is necessary to start OAP with the following configuration:
+
+```yaml
+receiver-async-profiler:
+  selector: ${SW_RECEIVER_ASYNC_PROFILER:default}
+  default:
+```
+
+## Async Profiler Task with Analysis
+
+To use the Async Profiler feature, please follow these steps:
+
+1. **Create Async Profiler task**: Use the UI or CLI tool to create a task.
+2. **Wait agent collect data and upload**: Wait for Async Profiler to collect 
JFR data and report
+3. **Query task progress**: Query the progress of tasks, including analyzing 
successful and failed instances and task logs
+4. **Analyze the data**: Analyze the JFR data to determine where performance 
bottlenecks exist in the service.
+
+### Create Async Profiler task
+
+Create an Async Profiler task to notify some java-agent instances in the 
execution service to start Async Profiler for data collection.
+
+When creating a task, the following configuration fields are required:
+
+1. **serviceId**: Define the service to execute the task.
+2. **serviceInstanceIds**: Define which instances need to execute tasks.
+3. **duration**: Define the duration of this task (second).
+4. **events**: Define which event types this task needs to collect.
+5. **execArgs**: Other Async Profiler execution options, e.g. alloc=2k,lock=2s.
+
+When the Agent receives a Async Profiler task from OAP, it automatically 
generates a log to notify that the task has been acknowledged. The log contains 
the following field information:
+
+1. **Instance**: The name of the instance where the Agent is located.
+2. **Type**: Supports "NOTIFIED" and "EXECUTION_FINISHED" and 
"JFR_UPLOAD_FILE_TOO_LARGE_ERROR", "EXECUTION_TASK_ERROR", with the current log 
displaying "NOTIFIED".
+3. **Time**: The time when the Agent received the task.
+
+### Wait agent collect data and upload
+
+At this point, async-profiler will trace the following kinds of events:   
+
+1. CPU cycles
+2. Hardware and Software performance counters like cache misses, branch 
misses, page faults, context switches etc.
+3. Allocations in Java Heap
+4. Contented lock attempts, including both Java object monitors and 
ReentrantLocks
+
+Finally, java agent will upload the jfr file produced by async-profiler to the 
oap server for online performance analysis.
+
+### Query task progress

Review Comment:
   ```suggestion
   ### Query the profiling task progresses
   ```



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to