[ 
https://issues.apache.org/jira/browse/STORM-885?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15011663#comment-15011663
 ] 

Kyle Nusbaum edited comment on STORM-885 at 11/18/15 6:52 PM:
--------------------------------------------------------------

[~LongdaFeng]
I disagree with your assessment of this patch.

Addressing each of your numbered points:
1. I'm not sure what you mean by this. Will you please clarify?
2. Pacemaker may become a bottleneck in the system, yes. But in our internal 
testing, Nimbus itself becomes a bottleneck before Pacemaker. Furthermore, it 
is *not* very hard to do extension. I have a prototype here for Pacemaker HA 
that only adds a small number of lines of code, which I am planning on pushing 
out in the relatively near future.
3. The logic looks fairly straight-forward to me, and the pacemaker server file 
itself is only 239 lines long, with nearly 100 of those being the license, 
imports, and statistics helpers. The client is only 125 lines long. 

Pacemaker is also an optional component, and it's ready now. 
I see the benefit of a 'TopologyMaster', but I don't see the benefit in 
blocking the merge of Pacemaker at this point.


was (Author: knusbaum):
[~LongdaFeng]
I disagree with your assessment of this patch.

1. I'm not sure what you mean by this. Will you please clarify?
2. Pacemaker may become a bottleneck in the system, yes. But in our internal 
testing, Nimbus itself becomes a bottleneck before Pacemaker. Furthermore, it 
is *not* very hard to do extension. I have a prototype here for Pacemaker HA 
that only adds a small number of lines of code, which I am planning on pushing 
out in the relatively near future.
3. The logic looks fairly straight-forward to me, and the pacemaker server file 
itself is only 239 lines long, with nearly 100 of those being the license, 
imports, and statistics helpers. The client is only 125 lines long. 

Pacemaker is also an optional component, and it's ready now. 
I see the benefit of a 'TopologyMaster', but I don't see the benefit in 
blocking the merge of Pacemaker at this point.

> Heartbeat Server (Pacemaker)
> ----------------------------
>
>                 Key: STORM-885
>                 URL: https://issues.apache.org/jira/browse/STORM-885
>             Project: Apache Storm
>          Issue Type: Improvement
>          Components: storm-core
>            Reporter: Robert Joseph Evans
>            Assignee: Kyle Nusbaum
>
> Large highly connected topologies and large clusters write a lot of data into 
> ZooKeeper.  The heartbeats, that make up the majority of this data, do not 
> need to be persisted to disk.  Pacemaker is intended to be a secure 
> replacement for storing the heartbeats without changing anything within the 
> heartbeats.  In the future as more metrics are added in, we may want to look 
> into switching it over to look more like Heron, where a metrics server is 
> running for each node/topology.  And can be used to aggregate/per-aggregate 
> them in a more scalable manor.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to