[
https://issues.apache.org/jira/browse/HADOOP-17461?focusedWorklogId=791114&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-791114
]
ASF GitHub Bot logged work on HADOOP-17461:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 14/Jul/22 19:49
Start Date: 14/Jul/22 19:49
Worklog Time Spent: 10m
Work Description: steveloughran commented on PR #4566:
URL: https://github.com/apache/hadoop/pull/4566#issuecomment-1184835595
you are going to love this. Full end to end stats collection of IO during
job commit of terasort.
```
2022-07-14 20:37:38,332 [JUnit] INFO s3a.AbstractS3ATestBase
(AbstractS3ATestBase.java:dumpFileSystemIOStatistics(137)) - Aggregate
FileSystem Statistics counters=((action_executor_acquired=2)
(action_file_opened=26)
(action_http_get_request=26)
(action_http_head_request=112)
(audit_request_execution=250)
(audit_span_creation=144)
(directories_created=14)
(directories_deleted=14)
(fake_directories_deleted=4)
(files_created=2)
(files_deleted=16)
(ignored_errors=8)
(object_bulk_delete_request=4)
(object_delete_objects=34)
(object_delete_request=14)
(object_list_request=74)
(object_metadata_request=112)
(object_put_request=16)
(object_put_request_completed=16)
(op_create=2)
(op_delete=16)
(op_exists=2)
(op_exists.failures=2)
(op_get_file_status=40)
(op_glob_status=4)
(op_list_files=6)
(op_list_located_status=4)
(op_list_status=4)
(op_list_status.failures=2)
(op_mkdirs=14)
(op_open=30)
(store_io_request=251)
(stream_read_bytes=232084)
(stream_read_close_operations=26)
(stream_read_closed=26)
(stream_read_fully_operations=10)
(stream_read_opened=26)
(stream_read_operations=41)
(stream_read_operations_incomplete=23)
(stream_read_remote_stream_drain=26)
(stream_read_seek_policy_changed=26)
(stream_read_total_bytes=232084)
(stream_write_block_uploads=2));
gauges=((stream_write_block_uploads_pending=2));
minimums=((action_executor_acquired.min=0)
(action_file_opened.min=49)
(action_http_get_request.min=63)
(action_http_head_request.min=40)
(object_bulk_delete_request.min=173)
(object_delete_request.min=51)
(object_list_request.min=48)
(object_put_request.min=82)
(op_create.min=62)
(op_delete.min=52)
(op_exists.failures.min=110)
(op_get_file_status.min=53)
(op_glob_status.min=102)
(op_list_files.min=176)
(op_list_status.failures.min=172)
(op_list_status.min=62)
(op_mkdirs.min=313)
(stream_read_remote_stream_drain.min=0));
maximums=((action_executor_acquired.max=0)
(action_file_opened.max=100)
(action_http_get_request.max=139)
(action_http_head_request.max=206)
(object_bulk_delete_request.max=454)
(object_delete_request.max=70)
(object_list_request.max=667)
(object_put_request.max=123)
(op_create.max=75)
(op_delete.max=685)
(op_exists.failures.max=120)
(op_get_file_status.max=178)
(op_glob_status.max=126)
(op_list_files.max=285)
(op_list_status.failures.max=184)
(op_list_status.max=68)
(op_mkdirs.max=847)
(stream_read_remote_stream_drain.max=1));
means=((action_executor_acquired.mean=(samples=2, sum=0, mean=0.0000))
(action_file_opened.mean=(samples=26, sum=1515, mean=58.2692))
(action_http_get_request.mean=(samples=26, sum=1997, mean=76.8077))
(action_http_head_request.mean=(samples=112, sum=7160, mean=63.9286))
(object_bulk_delete_request.mean=(samples=4, sum=997, mean=249.2500))
(object_delete_request.mean=(samples=14, sum=810, mean=57.8571))
(object_list_request.mean=(samples=74, sum=7647, mean=103.3378))
(object_put_request.mean=(samples=16, sum=1497, mean=93.5625))
(op_create.mean=(samples=2, sum=137, mean=68.5000))
(op_delete.mean=(samples=16, sum=1935, mean=120.9375))
(op_exists.failures.mean=(samples=2, sum=230, mean=115.0000))
(op_get_file_status.mean=(samples=40, sum=3283, mean=82.0750))
(op_glob_status.mean=(samples=4, sum=459, mean=114.7500))
(op_list_files.mean=(samples=6, sum=1280, mean=213.3333))
(op_list_status.failures.mean=(samples=2, sum=356, mean=178.0000))
(op_list_status.mean=(samples=2, sum=130, mean=65.0000))
(op_mkdirs.mean=(samples=14, sum=5303, mean=378.7857))
(stream_read_remote_stream_drain.mean=(samples=26, sum=3, mean=0.1154)));
Process finished with exit code 0
```
Issue Time Tracking
-------------------
Worklog Id: (was: 791114)
Time Spent: 4h 20m (was: 4h 10m)
> Add thread-level IOStatistics Context
> -------------------------------------
>
> Key: HADOOP-17461
> URL: https://issues.apache.org/jira/browse/HADOOP-17461
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs, fs/azure, fs/s3
> Affects Versions: 3.3.1
> Reporter: Steve Loughran
> Assignee: Mehakmeet Singh
> Priority: Major
> Labels: pull-request-available
> Time Spent: 4h 20m
> Remaining Estimate: 0h
>
> For effective reporting of the iostatistics of individual worker threads, we
> need a thread-level context which IO components update.
> * this contact needs to be passed in two background thread forming work on
> behalf of a task.
> * IO Components (streams, iterators, filesystems) need to update this context
> statistics as they perform work
> * Without double counting anything.
> I imagine a ThreadLocal IOStatisticContext which will be updated in the
> FileSystem API Calls. This context MUST be passed into the background threads
> used by a task, so that IO is correctly aggregated.
> I don't want streams, listIterators &c to do the updating as there is more
> risk of double counting. However, we need to see their statistics if we want
> to know things like "bytes discarded in backwards seeks". And I don't want to
> be updating a shared context object on every read() call.
> If all we want is store IO (HEAD, GET, DELETE, list performance etc) then the
> FS is sufficient.
> If we do want the stream-specific detail, then I propose
> * caching the context in the constructor
> * updating it only in close() or unbuffer() (as we do from S3AInputStream to
> S3AInstrumenation)
> * excluding those we know the FS already collects.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: common-issues-h...@hadoop.apache.org
