feat: Improve performance tracing feature (#1730)

Author: andygrove

docs/source/contributor-guide/tracing.md

@@ -19,18 +19,20 @@ under the License.
 
 # Tracing
 
-Tracing can be enabled by setting `spark.comet.tracing.enabled=true`. Additionally, enabling the `jemalloc` 
-feature will enable tracing of native memory allocations.
+Tracing can be enabled by setting `spark.comet.tracing.enabled=true`.
+
+With this feature enabled, each Spark executor will write a JSON event log file in
+Chrome's [Trace Event Format]. The file will be written to the executor's current working
+directory with the filename `comet-event-trace.json`.
+
+[Trace Event Format]: https://docs.google.com/document/d/1CvAClvFfyA5R-PhYUmn5OOQtYMH4h6I0nSsKchNAySU/preview?tab=t.0#heading=h.yr4qxyxotyw
+
+Additionally, enabling the `jemalloc` feature will enable tracing of native memory allocations.
 
 ```shell
 make release COMET_FEATURES="jemalloc"
 ```
 
-With this feature enabled, each Spark executor will write a JSON event log file in 
-Chrome's [Trace Event Format].
-
-[Trace Event Format]: https://docs.google.com/document/d/1CvAClvFfyA5R-PhYUmn5OOQtYMH4h6I0nSsKchNAySU/preview?tab=t.0#heading=h.yr4qxyxotyw
-
 Example output:
 
 ```json
@@ -43,10 +45,20 @@ Example output:
 { "name": "CometExecIterator_getNextBatch", "cat": "PERF", "ph": "B", "pid": 1, "tid": 5, "ts": 10109479156 },
 ```
 
-Traces can be viewed with [Trace Viewer]. 
+Traces can be viewed with [Trace Viewer].
 
 [Trace Viewer]: https://github.com/catapult-project/catapult/blob/main/tracing/README.md
 
 Example trace visualization:
 
 ![tracing](../_static/images/tracing.png)
+
+## Definition of Labels
+
+| Label                 | Meaning                                                        |
+| --------------------- | -------------------------------------------------------------- |
+| jvm_heapUsed          | JVM heap memory usage of live objects for the executor process |
+| jemalloc_allocated    | Native memory usage for the executor process                   |
+| task_memory_comet_NNN | Off-heap memory allocated by Comet for query execution         |
+| task_memory_spark_NNN | On-heap & Off-heap memory allocated by Spark                   |
+| comet_shuffle_NNN     | Off-heap memory allocated by Comet for columnar shuffle        |