Skip to content

[native] Add memory reclaim stats to task runtime stats #21794

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
xiaoxmeng merged 1 commit into from
Jan 26, 2024
Merged

[native] Add memory reclaim stats to task runtime stats #21794

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 15 additions & 2 deletions presto-native-execution/presto_cpp/main/PrestoTask.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -440,9 +440,22 @@ protocol::TaskInfo PrestoTask::updateInfoLocked() {

std::unordered_map<std::string, RuntimeMetric> taskRuntimeStats;

if (taskStats.memoryReclaimCount > 0) {
Copy link
Contributor

@mbasmanova mbasmanova Jan 26, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It would be nice to report these stats live while query is running.

There is code later that calls processTaskStats which clears taskRuntimeStats if query is not running.

Also, I noticed that Coordinator UI shows stats based on unit. Hence, you need to specify kNanos unit to ensure nice display.

For example,

  taskRuntimeStats.insert(
      {"nativeProcessCpuTime",
       RuntimeMetric(processCpuTime_, RuntimeCounter::Unit::kNanos)});

Copy link
Contributor Author

@xiaoxmeng xiaoxmeng Jan 26, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You mean to call processTaskStats with tryToSkipIfRunning to false?

taskRuntimeStats["memoryReclaimCount"].addValue(
taskStats.memoryReclaimCount);
taskRuntimeStats.insert(
{"memoryReclaimWallNanos",
RuntimeMetric(
taskStats.memoryReclaimMs * 1'000'000,
RuntimeCounter::Unit::kNanos)});
}

if (taskStats.endTimeMs >= taskStats.executionEndTimeMs) {
taskRuntimeStats["outputConsumedDelayInNanos"].addValue(
(taskStats.endTimeMs - taskStats.executionEndTimeMs) * 1'000'000);
taskRuntimeStats.insert(
{"outputConsumedDelayInNanos",
RuntimeMetric(
(taskStats.endTimeMs - taskStats.executionEndTimeMs) * 1'000'000,
RuntimeCounter::Unit::kNanos)});
taskRuntimeStats["createTime"].addValue(taskStats.executionStartTimeMs);
taskRuntimeStats["endTime"].addValue(taskStats.endTimeMs);
}
Expand Down