[
https://issues.apache.org/jira/browse/HADOOP-19357?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17904021#comment-17904021
]
ASF GitHub Bot commented on HADOOP-19357:
-----------------------------------------
manika137 opened a new pull request, #7216:
URL: https://github.com/apache/hadoop/pull/7216
### Description of PR
JIRA: https://issues.apache.org/jira/browse/HADOOP-19357
Given the significant improvements on the service side over the past few
years, the current client-side retry mechanism, which enforces 5-6 minutes of
sleep time with jitter, is now excessive. Therefore, we are realigning the
retry and throttling logic thresholds.
Since the service can now handle higher traffic rates, we aim to enhance
driver performance by setting the following configurations as default:
**Client-side throttling (CST):** Off
**Client Backoff -** 500ms
**Max Backoff -** 25s
**Min Backoff -** 500ms
## The test results are as follows:
============================================================
HNS-OAuth
============================================================
[WARNING] Tests run: 157, Failures: 0, Errors: 0, Skipped: 3
[WARNING] Tests run: 645, Failures: 0, Errors: 0, Skipped: 85
[WARNING] Tests run: 171, Failures: 0, Errors: 0, Skipped: 25
[WARNING] Tests run: 262, Failures: 0, Errors: 0, Skipped: 32
============================================================
HNS-SharedKey
============================================================
[WARNING] Tests run: 157, Failures: 0, Errors: 0, Skipped: 4
[WARNING] Tests run: 648, Failures: 0, Errors: 0, Skipped: 35
[WARNING] Tests run: 171, Failures: 0, Errors: 0, Skipped: 25
[WARNING] Tests run: 262, Failures: 0, Errors: 0, Skipped: 19
============================================================
NonHNS-SharedKey
============================================================
[WARNING] Tests run: 157, Failures: 0, Errors: 0, Skipped: 10
[WARNING] Tests run: 632, Failures: 0, Errors: 0, Skipped: 275
[WARNING] Tests run: 171, Failures: 0, Errors: 0, Skipped: 27
[WARNING] Tests run: 262, Failures: 0, Errors: 0, Skipped: 20
============================================================
AppendBlob-HNS-OAuth
============================================================
[WARNING] Tests run: 157, Failures: 0, Errors: 0, Skipped: 3
[WARNING] Tests run: 653, Failures: 0, Errors: 0, Skipped: 87
[WARNING] Tests run: 171, Failures: 0, Errors: 0, Skipped: 49
[WARNING] Tests run: 262, Failures: 0, Errors: 0, Skipped: 32
> ABFS: Optimizations for Retry Handling and Throttling
> -----------------------------------------------------
>
> Key: HADOOP-19357
> URL: https://issues.apache.org/jira/browse/HADOOP-19357
> Project: Hadoop Common
> Issue Type: Improvement
> Components: fs/azure
> Affects Versions: 3.4.0, 3.4.1
> Reporter: Manika Joshi
> Assignee: Manika Joshi
> Priority: Minor
>
> Given the significant improvements on the service side over the past few
> years, the current client-side retry mechanism, which enforces 5-6 minutes of
> sleep time with jitter, is now excessive. Therefore, we are realigning the
> retry and throttling logic thresholds.
> Since the service can now handle higher traffic rates, we aim to enhance
> driver performance by setting the following configurations as default:
> * {*}Client-side throttling (CST){*}: Off
> * *Client Backoff* - 500ms
> * *Max Backoff* - 25s
> * *Min Backoff* - 500ms
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
