Hello Benjamin,
How about this patch?
Look forward your replies.
Thank you very much.
Best regards!
On 2017/5/19 17:45, Yang Feng wrote:
> Hi Benjamin,
>
> Thank you very much for your comments.
> Please find my replys and the up-to-date patch.
> Best regards!
>
>>
>> First, one overall que
Hello Benjamin and Christophe,
How about this patch?
Thanks a lot.
Best.
On 2017/5/19 17:45, Yang Feng wrote:
> Hi Benjamin,
>
> Thank you very much for your comments.
> Please find my replys and the up-to-date patch.
> Best regards!
>
>>
>> First, one overall question. We have dynamic path s
Hello Martin and Christophe,
How about this patch?
Thanks a lot.
Best.
On 2017/5/19 16:43, Yang Feng wrote:
> Hello Martin,
>
> Firstly, thank you very much for your comments.
> And find my replys and the up-to-date patch.
>
> Best regards!
>
>
>> Please think about the name once again. Mayb
Hi Benjamin,
Thank you very much for your comments.
Please find my replys and the up-to-date patch.
Best regards!
>
> First, one overall question. We have dynamic path selectors available to
> deal with paths that are just simply slower that other paths, but can
> still be used together. Is the
Hello Martin,
Firstly, thank you very much for your comments.
And find my replys and the up-to-date patch.
Best regards!
> Please think about the name once again. Maybe you should call it
> "io_latency" or "path_latency" instead of "delayedpath"?
OK, as the following patch.
>
> Hm, I can't se
t: [PATCH] [dm-devel] [PATCH] multipath-tools:Prioritizer based on a
> time-delay algorithm
> Prioritizer for device mapper multipath, where the corresponding priority
> values of specific paths are provided by a time-delay
> algorithm. And the time-delay algorithm is dependent on the fo
> > +avgdelay = toldelay/(long long)(cons_num - 2);
> > > +if (avgdelay > THRES_USEC_VALUE)
> > > +{
> > > + condlog(0, "%s: avgdelay is more than thresold", pp-
> > > >dev);
> > > +retur
as the following patch.
>
>>> + toldelay -= min + max;
>>> +avgdelay = toldelay/(long long)(cons_num - 2);
>>> +if (avgdelay > THRES_USEC_VALUE)
>>> +{
>>> +condlog(0, "%s: avgdelay is more than thres
return 1;
>> +}
>> +
>> +ratio = get_conversion_ratio(type);
>> +rc = (int)(THRES_USEC_VALUE - (avgdelay/(((long
>> long)delay_interval) * ratio)));
>> +
>> + return rc;
>> +}
>
> Is it reasonable to do these inter
Hello Yang,
thank you for your work. Please find my remarks below.
On Mon, 2017-05-08 at 11:58 +0800, Yang Feng wrote:
> Prioritizer for device mapper multipath, where the corresponding
> priority
> values of specific paths are provided by a time-delay algorithm. And
> the
> time-delay algorithm
Hi Xose and Christophe,
On 2017/5/11 6:36, Xose Vazquez Perez wrote:
> On 05/08/2017 05:58 AM, Yang Feng wrote:
>
>> Prioritizer for device mapper multipath, where the corresponding priority
>> values of specific paths are provided by a time-delay algorithm. And the
>> time-delay algorithm is depe
On 05/08/2017 05:58 AM, Yang Feng wrote:
> Prioritizer for device mapper multipath, where the corresponding priority
> values of specific paths are provided by a time-delay algorithm. And the
> time-delay algorithm is dependent on the following arguments(delay_interval,
> cons_num).
This new featu
Prioritizer for device mapper multipath, where the corresponding priority
values of specific paths are provided by a time-delay algorithm. And the
time-delay algorithm is dependent on the following arguments(delay_interval,
cons_num).
The principle of the algorithm is illustrated as follows:
1. By
13 matches
Mail list logo
