[ 
https://issues.apache.org/jira/browse/YARN-3409?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15686934#comment-15686934
 ] 

Varun Saxena commented on YARN-3409:
------------------------------------

Will take the liberty of answering few of these questions.

bq. Nodes with same Constraints could be in multiple partition,so are we 
allowing resource to serve any labels satisfying constraints or only subset of 
label resource pool satisfying constraints.
Constraints will be matched within the scope of a node partition.

bq. We could also support default constraints DISK,CPUS,JAVA version etc.Are we 
planning to report from Nodemanager these default constraints. 
Plan to have these things in future. We can either have a separate interface or 
 use the script based node labels provider interface. 

bq. IIUC, currently headroom is calculated per partition basis, how does head 
room will be calculated once we support constraint label? What is the impact ? 
Does it application wait for ever if required constraint label not found?
Constraints will merely be used for node selection.
Application can update ask and specify different constraints (or none at all) 
if its fine for it to launch containers elsewhere and its not getting the nodes 
matching the constraints since a long time. RM won't have something equivalent 
to locality delay here.

> Add constraint node labels
> --------------------------
>
>                 Key: YARN-3409
>                 URL: https://issues.apache.org/jira/browse/YARN-3409
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: api, capacityscheduler, client
>            Reporter: Wangda Tan
>            Assignee: Naganarasimha G R
>         Attachments: Constraint-Node-Labels-Requirements-Design-doc_v1.pdf
>
>
> Specify only one label for each node (IAW, partition a cluster) is a way to 
> determinate how resources of a special set of nodes could be shared by a 
> group of entities (like teams, departments, etc.). Partitions of a cluster 
> has following characteristics:
> - Cluster divided to several disjoint sub clusters.
> - ACL/priority can apply on partition (Only market team / marke team has 
> priority to use the partition).
> - Percentage of capacities can apply on partition (Market team has 40% 
> minimum capacity and Dev team has 60% of minimum capacity of the partition).
> Constraints are orthogonal to partition, they’re describing attributes of 
> node’s hardware/software just for affinity. Some example of constraints:
> - glibc version
> - JDK version
> - Type of CPU (x86_64/i686)
> - Type of OS (windows, linux, etc.)
> With this, application can be able to ask for resource has (glibc.version >= 
> 2.20 && JDK.version >= 8u20 && x86_64).



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

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

Reply via email to