samza-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Yi Pan (Data Infrastructure)" <yi...@linkedin.com>
Subject Re: Review Request 40626: SAMZA-668 - Document YARN host affinity in the website
Date Tue, 24 Nov 2015 18:06:52 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/40626/#review107807
-----------------------------------------------------------

Ship it!


LGTM. Found a few nits. Feel free to commit after fix them. Thanks!


docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 26)
<https://reviews.apache.org/r/40626/#comment167085>

    nit: "and resubmitted."



docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 30)
<https://reviews.apache.org/r/40626/#comment167086>

    nit: needs to *be* restored every time a container *starts up*.



docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 38)
<https://reviews.apache.org/r/40626/#comment167087>

    nit: trailing white space.



docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 43)
<https://reviews.apache.org/r/40626/#comment167088>

    Why is there a trailing '/*' here?



docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 50)
<https://reviews.apache.org/r/40626/#comment167089>

    trailing white space.



docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 63)
<https://reviews.apache.org/r/40626/#comment167092>

    nit: as we no longer *consume* from



docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 101)
<https://reviews.apache.org/r/40626/#comment167095>

    prefer: Configuring a Samza job *to use* Host Affinity



docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 107)
<https://reviews.apache.org/r/40626/#comment167097>

    nit: should *not have* any adverse effect



docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 111)
<https://reviews.apache.org/r/40626/#comment167098>

    nit: due to *variable* load distribution ... Hence, this *is a best-effort policy* that
Samza provides.
    
    nit: these guarantees may be hard to achieve or *are* not applicable.



docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 113)
<https://reviews.apache.org/r/40626/#comment167099>

    nit: there is no logic in the *Job Coordinator* to handle



docs/learn/documentation/versioned/yarn/yarn-host-affinity.md (line 114)
<https://reviews.apache.org/r/40626/#comment167100>

    prefer: the grouper logic used *to distribute* the partitions across...
    nit: the data in the *CoordinatorStream (for changelog-task partition assignment etc)*
and the data...


- Yi Pan (Data Infrastructure)


On Nov. 24, 2015, 2:07 a.m., Navina Ramesh wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/40626/
> -----------------------------------------------------------
> 
> (Updated Nov. 24, 2015, 2:07 a.m.)
> 
> 
> Review request for samza, Boris Shkolnik, Yan Fang, Chris Riccomini, Jake Maes, Jagadish
Venkatraman, Xinyu Liu, and Yi Pan (Data Infrastructure).
> 
> 
> Repository: samza
> 
> 
> Description
> -------
> 
> Added documentation for host affinity feature in samza
> Config details are already available in configuration-table.html. Hence, you won't see
it in the diff.
> 
> 
> Diffs
> -----
> 
>   docs/img/versioned/learn/documentation/yarn/samza-host-affinity.png PRE-CREATION 
>   docs/learn/documentation/versioned/index.html dec5be15453a8e5794a9c66b0010803f591ed0cf

>   docs/learn/documentation/versioned/yarn/application-master.md 807f6f312f2361cecdd4ae9911bb5cf59115ae85

>   docs/learn/documentation/versioned/yarn/yarn-host-affinity.md PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/40626/diff/
> 
> 
> Testing
> -------
> 
> Locally deployed the site and tested
> 
> 
> Thanks,
> 
> Navina Ramesh
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message