[jira] [Commented] (AMBARI-3015) Ambari API latency increases linearly to load

["Srimanth Gunturi (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/AMBARI-3015?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13752567#comment-13752567 ] 

Srimanth Gunturi commented on AMBARI-3015:
------------------------------------------

This might not be an Ambari issue after all. When I attempted the same on physically different machines, I did not see the time increase. I attempted the above on VMs, and they might have been allocated on a single processor resulting in linear increase with load. 
                
> Ambari API latency increases linearly to load
> ---------------------------------------------
>
>                 Key: AMBARI-3015
>                 URL: https://issues.apache.org/jira/browse/AMBARI-3015
>             Project: Ambari
>          Issue Type: Bug
>    Affects Versions: 1.4.0
>            Reporter: Srimanth Gunturi
>            Assignee: Mahadev konar
>              Labels: performance
>             Fix For: 1.4.0
>
>         Attachments: 1Thread.png, 2Threads.png, 3Threads.png
>
>
> Setup: In a thread I am calling Ambari UI's 5 most used APIs. They are:
> {noformat}
> /api/v1/clusters/${cluster}/hosts?fields=Hosts/host_name,Hosts/public_host_name,Hosts/disk_info,Hosts/cpu_count,Hosts/total_mem,Hosts/host_status,Hosts/last_heartbeat_time,Hosts/os_arch,Hosts/os_type,Hosts/ip,host_components,metrics/disk,metrics/load/load_one
> {noformat}
> {noformat}
> /api/v1/clusters/${cluster}/services?fields=components/ServiceComponentInfo,components/host_components,components/host_components/HostRoles,components/host_components/metrics/jvm/memHeapUsedM,components/host_components/metrics/jvm/memHeapCommittedM,components/host_components/metrics/mapred/jobtracker/trackers_decommissioned,components/host_components/metrics/cpu/cpu_wio,components/host_components/metrics/rpc/RpcQueueTime_avg_time,components/host_components/metrics/flume/flume,components/host_components/metrics/yarn/Queue
> {noformat}
> {noformat}
> /api/v1/clusters/vm/requests?to=end&page_size=10&fields=*,tasks/Tasks/*
> {noformat}
> {noformat}
> /api/v1/clusters/vm/host_components?fields=HostRoles/nagios_alerts&HostRoles/component_name=NAGIOS_SERVER
> {noformat}
> {noformat}
> /api/v1/clusters/vm/services?fields=ServiceInfo,components/host_components/HostRoles/desired_state,components/host_components/HostRoles/state
> {noformat}
> The above I call one after the other and loop 100 times.
> When I run 2 threads in parallel, it takes twice as long as 1 thread. 
> When I run 3 threads in parallel, it takes thrice as long as 1 thread.
> It looks like each sessions requests are done one after the other in serial.
> Images attached. 

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira