[Gluster-devel] [Gluster-users] What's the impact of enabling the profiler?

Kaushal M kshlmster at gmail.com
Tue Jul 22 08:16:43 UTC 2014


Joe,
BTW, Vipul (in CC) is also looking to quantify the impact of enabling
io-stats counters permanently.
It'd be great if you both could help each other out.

~kaushal

On Tue, Jul 22, 2014 at 12:54 PM, Pranith Kumar Karampuri
<pkarampu at redhat.com> wrote:
>
> On 07/22/2014 12:07 PM, Kaushal M wrote:
>>
>> The clear you have at top (the one for "clear-stats") is for 'top
>> clear'. Below there is a section with GF_CLI_INFO_CLEAR, which handles
>> 'profile info clear'.
>
> Cool. Thanks :-). The stuff that is present on master for 'profile' is so
> much better :-)
>
> Pranith
>
>>
>> ~kaushal
>>
>> On Tue, Jul 22, 2014 at 11:50 AM, Pranith Kumar Karampuri
>> <pkarampu at redhat.com> wrote:
>>>
>>> On 07/22/2014 11:43 AM, Kaushal M wrote:
>>>>
>>>> 'gluster volume profile <VOLNAME> info clear' should clear the stats as
>>>> well.
>>>>
>>>>   From 'gluster volume help'
>>>>     volume profile <VOLNAME> {start|info [peek|incremental
>>>> [peek]|cumulative|clear]|stop} [nfs]
>>>
>>> According to the code it is clearing only stats of 'top' not of the
>>> 'profile'
>>>
>>> Pranith
>>>
>>>> On Tue, Jul 22, 2014 at 11:39 AM, Joe Julian <joe at julianfamily.org>
>>>> wrote:
>>>>>
>>>>> On 07/17/2014 07:30 PM, Pranith Kumar Karampuri wrote:
>>>>>>
>>>>>>
>>>>>> On 07/18/2014 03:05 AM, Joe Julian wrote:
>>>>>>>
>>>>>>> What impact, if any, does starting profiling (gluster volume profile
>>>>>>> $vol
>>>>>>> start) have on performance?
>>>>>>
>>>>>> Joe,
>>>>>>       According to the code the only extra things it does is calling
>>>>>> gettimeofday() call at the beginning and end of the FOP to calculate
>>>>>> latency, increment some variables. So I guess not much?
>>>>>>
>>>>> So far so good. Is the only way to clear the stats to restart the
>>>>> brick?
>>>>>
>>>>> _______________________________________________
>>>>> Gluster-devel mailing list
>>>>> Gluster-devel at gluster.org
>>>>> http://supercolony.gluster.org/mailman/listinfo/gluster-devel
>>>
>>>
>


More information about the Gluster-devel mailing list