[
https://issues.apache.org/jira/browse/HDFS-2832?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13707422#comment-13707422
]
Sanjay Radia edited comment on HDFS-2832 at 7/12/13 10:21 PM:
--------------------------------------------------------------
Andrew, I was always confused about the creation of HDFS-4672 given that
HDFS-2832 existed for quite a while and with a bunch of subtasks complete.
note: Konstantine asked you about the duplication in April "What is the
difference between this issue and HDFS-2832?" in [comment |
https://issues.apache.org/jira/browse/HDFS-4672?focusedCommentId=13629403&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13629403]
to which you have not responded to date.
However noting the description and comments on HDFS-4672 it is clear that
HDFS-4672 seems to be focusing on policies for placement. I resigned to letting
this Jira (HDFS-2832) focus on the raw mechanisms and HDFS-4672 focus on the
policies; as a result I did not raise the issue about you duplicating this jira
when you created HDFS-4672. I did that out of politeness and moving forward but
perhaps that was a mistake on my part. Also note that I have commented on the
other jira and have not ignored it.
Your recent tweet about this before giving me a chance to respond is
unfortunate, especially given that HDFS-4672 is overlapping with HDFS-2832. -
isn't this the pot calling the kettle black?
Getting back to the business building Hadoop:
We need to move quickly forward by adding mechanism in HDFS to allow different
kinds of storage while the longer discussions on policies continue. Given that
a recent Jira on ram caching (HDFS-4949) is planning to move quickly, it is
important that we start by providing the the needed mechanisms. We plan to
add additional mechanisms via sub-jiras which can support HDFS-4949. You are
welcome to contribute with patches and discussions. Similarly we plan to
continue to participate in the longer ranged discussions on policies and also
contribute with patches.
was (Author: sanjay.radia):
Andrew, I was always confused about the creation of HDFS-4672 given that
HDFS-2832 existed for quite a while and with a bunch of subtasks complete.
note: Konstantine asked you about the duplication in April "What is the
difference between this issue and HDFS-2832?" in [comment |
https://issues.apache.org/jira/browse/HDFS-4672?focusedCommentId=13629403&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13629403]
to which you have not responded to date.
However noting the description and comments on HDFS-4672 it is clear that
HDFS-4672 seems to be focusing on policies for placement. I resigned to letting
this Jira (HDFS-2832) focus on the raw mechanisms and HDFS-4672 focus on the
policies; as a result I did not raise the issue about you duplicating this jira
when you created HDFS-4672. I did that out of politeness and moving forward but
perhaps that was a mistake on my part. Also note that I have commented on the
other jira and have not ignored it.
Your recent tweet about this before giving me a chance to respond is
unfortunate, especially given that HDFS-4682 has is overlapping with HDFS-2832.
- isn't this the pot calling the kettle black?
Getting back to the business building Hadoop:
We need to move quickly forward by adding mechanism in HDFS to allow different
kinds of storage while the longer discussions on policies continue. Given that
a recent Jira on ram caching (HDFS-4949) is planning to move quickly, it is
important that we start by providing the the needed mechanisms. We plan to
add additional mechanisms via sub-jiras which can support HDFS-4949. You are
welcome to contribute with patches and discussions. Similarly we plan to
continue to participate in the longer ranged discussions on policies and also
contribute with patches.
> Enable support for heterogeneous storages in HDFS
> -------------------------------------------------
>
> Key: HDFS-2832
> URL: https://issues.apache.org/jira/browse/HDFS-2832
> Project: Hadoop HDFS
> Issue Type: New Feature
> Affects Versions: 0.24.0
> Reporter: Suresh Srinivas
> Assignee: Suresh Srinivas
>
> HDFS currently supports configuration where storages are a list of
> directories. Typically each of these directories correspond to a volume with
> its own file system. All these directories are homogeneous and therefore
> identified as a single storage at the namenode. I propose, change to the
> current model where Datanode * is a * storage, to Datanode * is a collection
> * of strorages.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
