hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Íñigo Goiri (JIRA) <j...@apache.org>
Subject [jira] [Commented] (HDFS-14090) RBF: Improved isolation for downstream name nodes.
Date Thu, 01 Aug 2019 17:57:00 GMT

    [ https://issues.apache.org/jira/browse/HDFS-14090?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16898235#comment-16898235

Íñigo Goiri commented on HDFS-14090:

Thanks [~crh] for the update.
I had the chance to go a little deeper and I have a concern about the "Permit" wording.
It is "followable" but I'm not sure is the most common terminology.
{{PermitAllocationException}} is like a setup exception so something closer to {{IllegalArgumentException}}.
{{NoPermitAvailableException}} is something like too many requests, not sure what the best
mapping here would be.
Eventually it surfaces as the Router is overloaded which is good, however for code readability
we may want to use a more intuitive concept throughout the code.
[~xkrogen], you have been working on the RPC fairness, do you have any suggestion for the

Other minor comments:
* logFinalAssignment() is a one-liner that could just be there directly. logAssignment() at
least is used a few more times.
* {{LOG.info("Final permit allocation table {}", this.permits.toString());}} no need for toString().
* Instead of having two configs, one for enabling and one for the implementation, we could
have just the implementation and by default provide a dummy implementation that doesn't do
fairness. Then we would rename the current DefaultFairnessPolicyController to something more
descriptive (to reflect equal or linear or similar).
* Coming back to {{PermitAllocationException}}, right now we are kind of logging and swallowing;
what about failing the whole startup?
* {{TestRouterFairnessManager#160}} could just use a for loop.

> RBF: Improved isolation for downstream name nodes.
> --------------------------------------------------
>                 Key: HDFS-14090
>                 URL: https://issues.apache.org/jira/browse/HDFS-14090
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: CR Hota
>            Assignee: CR Hota
>            Priority: Major
>         Attachments: HDFS-14090-HDFS-13891.001.patch, HDFS-14090-HDFS-13891.002.patch,
HDFS-14090-HDFS-13891.003.patch, HDFS-14090-HDFS-13891.004.patch, HDFS-14090-HDFS-13891.005.patch,
HDFS-14090.006.patch, HDFS-14090.007.patch, HDFS-14090.008.patch, HDFS-14090.009.patch, RBF_
Isolation design.pdf
> Router is a gateway to underlying name nodes. Gateway architectures, should help minimize
impact of clients connecting to healthy clusters vs unhealthy clusters.
> For example - If there are 2 name nodes downstream, and one of them is heavily loaded
with calls spiking rpc queue times, due to back pressure the same with start reflecting on
the router. As a result of this, clients connecting to healthy/faster name nodes will also
slow down as same rpc queue is maintained for all calls at the router layer. Essentially the
same IPC thread pool is used by router to connect to all name nodes.
> Currently router uses one single rpc queue for all calls. Lets discuss how we can change
the architecture and add some throttling logic for unhealthy/slow/overloaded name nodes.
> One way could be to read from current call queue, immediately identify downstream name
node and maintain a separate queue for each underlying name node. Another simpler way is to
maintain some sort of rate limiter configured for each name node and let routers drop/reject/send
error requests after certain threshold. 
> This won’t be a simple change as router’s ‘Server’ layer would need redesign
and implementation. Currently this layer is the same as name node.
> Opening this ticket to discuss, design and implement this feature.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org

View raw message