[
https://issues.apache.org/jira/browse/AMBARI-17376?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15344741#comment-15344741
]
Srimanth Gunturi commented on AMBARI-17376:
-------------------------------------------
Took some raw data measurements for *five* /recommendations call on 900 node
cluster for recommending configurations:
{noformat}
Client Call Time 2790 2828 2595 2550 2601
network,serialization,.. 803 1003 755 659 760
getHostsInformation 362 282 315 310 286
getServicesInformation 153 155 138 152 162
loadJson(hosts): 1012 922 938 972 943
loadJson(services) 216 207 208 212 204
instantiateStackAdvisor() 22 25 21 21 21
recommendConfigurations(): 171 187 171 174 174
dumpJson(): 5 4 4 4 4
{noformat}
>From the information above, approx 36% of entire client time is being spent on
>loading just the _hosts.json_ by Python. We definitely need to speed up
>loading JSON files in Python to speed up the most. The actual recommendations
>logic takes only ~173ms.
!RecommendationsBreakdown.png!
> Recommendations request execution takes too long time
> -----------------------------------------------------
>
> Key: AMBARI-17376
> URL: https://issues.apache.org/jira/browse/AMBARI-17376
> Project: Ambari
> Issue Type: Bug
> Components: ambari-server
> Affects Versions: 2.4.0
> Reporter: Srimanth Gunturi
> Assignee: Srimanth Gunturi
> Priority: Critical
> Fix For: 2.4.0
>
>
> /api/v1/stacks/HDP/versions/2.5/recommendations GET request is executed for a
> long time.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
