> On Jan. 13, 2016, 1:52 a.m., Guangya Liu wrote:
> > Since the comments are mainly to developers, the `second stage` is not very 
> > clear to developers, but the `WDRF` seems more clear. If want to update, 
> > what about "fair share stage"?

The fairness of the "second" stage is determined by sorters as well. For 
example, one may write a sorter that favors roles starting with "alex-" : ). 
What allocator knows is that there are two stages: the first where only 
quota'ed frameworks are processed and the second for all frameworks.


- Alexander


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/42221/#review114126
-----------------------------------------------------------


On Jan. 12, 2016, 11:06 p.m., Alexander Rukletsov wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/42221/
> -----------------------------------------------------------
> 
> (Updated Jan. 12, 2016, 11:06 p.m.)
> 
> 
> Review request for mesos, Ben Mahler and Joris Van Remoortere.
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Since wDRF is encapsulated in the choice of sorter, it should not be
> mentioned in the allocator code.
> 
> 
> Diffs
> -----
> 
>   src/master/allocator/mesos/hierarchical.cpp 
> df8bccaf2b8cfc0cb5ca18d4867371ae7a84c12f 
> 
> Diff: https://reviews.apache.org/r/42221/diff/
> 
> 
> Testing
> -------
> 
> `make check` on Mac OS 10.10.4
> 
> 
> Thanks,
> 
> Alexander Rukletsov
> 
>

Reply via email to