-
Notifications
You must be signed in to change notification settings - Fork 51
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.
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
Why the blkio_delay is so different with iotop-py? #65
Comments
I found the logic, you added all threads statistics to main threads besides to thread itself. Seems the main thread is used for filter and ordering, thus it must sum all threads data; while corrupts the outputs of main thread |
Right, maybe the current behavior is erratic. I need to double check it anyways... The intent was to have thread data summed in In order to switch between both modes, What do you think? |
Yes, it should maintain the process level information independently. And I found another possible problem, And, I think IO should not be summed up at process level, maybe max(IO) or average(IO) of all threads is better. |
Thanks for spotting these problems! I will separate the information. But if there is After checking the code, I can confirm that it is possible The aggregation function currently in use should have been I think that adding a toggle to switch between To summarize: Bug 1) Do not aggregate the data in the main thread [thread mode only]; keep both aggregated main thread data and its original data to enable proper switching between thread and process mode. |
I made a busy reading of disk in clickhouse, and notice the GRAPH[IO]▽ column in the view is quite different with the IO> column in iotop.
The main thread of clickhouse shows the GRAPH[IO] is 100% and IO> is zero.
Which I think iotop is correct.
Because the IO mainly happens in a background thread, the main thread should be idle.
I see the data is fetched from netlink instead of /proc; but the algorithm is similar.
I don't understand the result, is there something I missed that GRAPH[IO] is not equivalent to IO> ?
The text was updated successfully, but these errors were encountered: