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 50174: SAMZA-977: User doc for samza multithreading
Date Wed, 07 Sep 2016 18:01:27 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/50174/#review148041
-----------------------------------------------------------



Thanks for the detailed doc. I mainly have question regarding to the event-loop.md. We can
talk offline if you don't agree/understand my comments.


docs/learn/documentation/versioned/api/overview.md (line 37)
<https://reviews.apache.org/r/50174/#comment215413>

    java NIO ==> Java NIO



docs/learn/documentation/versioned/api/overview.md (line 49)
<https://reviews.apache.org/r/50174/#comment215414>

    This is a bit confusing. If I read these two comment lines as pseudo code, I would think
that the same thread is going to call asynchronous calls and fire callback. Is it the case?
I thought that the callback is triggered in a callback thread, not in the main thread invoking
the asynchronous calls?



docs/learn/documentation/versioned/container/event-loop.md (line 26)
<https://reviews.apache.org/r/50174/#comment215418>

    What do you try to say in the "Without explicit user synchronization..."? Is it by design,
we are trying to make sure the global variables implements a read-after-write consistency
among tasks, w/ the atomic step being the task operation? P.S. what's exactly a task operation?
It is better to define what this is.



docs/learn/documentation/versioned/container/event-loop.md (line 28)
<https://reviews.apache.org/r/50174/#comment215420>

    This section is a bit confusing to me. The description above in "Without..." sounds like
that we are implementing read-after-write consistency on global variables. Here we are saying
that global variables have to implement the sync method by the user.



docs/learn/documentation/versioned/container/event-loop.md (line 36)
<https://reviews.apache.org/r/50174/#comment215423>

    So, this describes the AsyncRunLoop logic. Don't we still have the synchronous RunLoop
in SamzaContainer?



docs/learn/documentation/versioned/container/event-loop.md (line 50)
<https://reviews.apache.org/r/50174/#comment215426>

    from *a* different user thread.


- Yi Pan (Data Infrastructure)


On Sept. 7, 2016, 5:16 p.m., Xinyu Liu wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/50174/
> -----------------------------------------------------------
> 
> (Updated Sept. 7, 2016, 5:16 p.m.)
> 
> 
> Review request for samza, Chris Pettitt, Navina Ramesh, and Yi Pan (Data Infrastructure).
> 
> 
> Repository: samza
> 
> 
> Description
> -------
> 
> Update samza web docs with new multithreading api, core and configs.
> 
> 
> Diffs
> -----
> 
>   docs/learn/documentation/versioned/api/overview.md 6712344e84e19883b857e00549db2acb101c7e0e

>   docs/learn/documentation/versioned/container/event-loop.md 116238312df7071747cbbc14bc9c46f558755195

>   docs/learn/documentation/versioned/jobs/configuration-table.html 54c52981c3055b398ee60af50eeaf2592ed0e64f

> 
> Diff: https://reviews.apache.org/r/50174/diff/
> 
> 
> Testing
> -------
> 
> Test the web pages locally.
> 
> 
> Thanks,
> 
> Xinyu Liu
> 
>


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