You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Grant Ingersoll (Created) (JIRA)" <ji...@apache.org> on 2012/02/22 20:47:49 UTC
[jira] [Created] (SOLR-3154) SolrJ CloudServer should be leader
aware when adding docs
SolrJ CloudServer should be leader aware when adding docs
---------------------------------------------------------
Key: SOLR-3154
URL: https://issues.apache.org/jira/browse/SOLR-3154
Project: Solr
Issue Type: Improvement
Components: SolrCloud
Affects Versions: 4.0
Reporter: Grant Ingersoll
Priority: Minor
Fix For: 4.0
It would be good when indexing if the SolrJ CloudServer was leader aware so that we could avoid doing an extra hop for the data.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Updated] (SOLR-3154) SolrJ CloudServer should be leader and
network aware when adding docs
Posted by "Mark Miller (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-3154?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Mark Miller updated SOLR-3154:
------------------------------
Fix Version/s: (was: 4.0)
> SolrJ CloudServer should be leader and network aware when adding docs
> ---------------------------------------------------------------------
>
> Key: SOLR-3154
> URL: https://issues.apache.org/jira/browse/SOLR-3154
> Project: Solr
> Issue Type: Improvement
> Components: SolrCloud
> Affects Versions: 4.0-ALPHA
> Reporter: Grant Ingersoll
> Assignee: Mark Miller
> Priority: Minor
> Fix For: 5.0
>
> Attachments: SOLR-3154.patch
>
>
> It would be good when indexing if the SolrJ CloudServer was leader aware so that we could avoid doing an extra hop for the data. It would also be good if one could easily set things up based on data locality principles. This might mean that CloudServer is aware of where on the network it is and would pick leaders that are as close as possible (i.e. local, perhaps.) This would come in to play when working with tools like Hadoop or other grid computing frameworks.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
Re: [jira] [Updated] (SOLR-3154) SolrJ CloudServer should be leader
and network aware when adding docs
Posted by Chris Hostetter <ho...@fucit.org>.
Wait a minute ... this fixVersion update caught my eye.
the 4.0-BETA release highlights said...
* Improved Solrj client performance with Solr Cloud: updates are
only sent to leaders by default.
...and i just merged that into the 4.0 final release highlights -- but
based on this issue description and fix version, i'm now wondering if that
was a mistake in the release announcement?
: Mark Miller updated SOLR-3154:
: ------------------------------
:
: Fix Version/s: 4.1
:
: > SolrJ CloudServer should be leader and network aware when adding docs
: > ---------------------------------------------------------------------
: >
: > Key: SOLR-3154
: > URL: https://issues.apache.org/jira/browse/SOLR-3154
: > Project: Solr
: > Issue Type: Improvement
: > Components: SolrCloud
: > Affects Versions: 4.0-ALPHA
: > Reporter: Grant Ingersoll
: > Assignee: Mark Miller
: > Priority: Minor
: > Fix For: 4.1, 5.0
: >
: > Attachments: SOLR-3154.patch
: >
: >
: > It would be good when indexing if the SolrJ CloudServer was leader aware so that we could avoid doing an extra hop for the data. It would also be good if one could easily set things up based on data locality principles. This might mean that CloudServer is aware of where on the network it is and would pick leaders that are as close as possible (i.e. local, perhaps.) This would come in to play when working with tools like Hadoop or other grid computing frameworks.
:
: --
: 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
:
: ---------------------------------------------------------------------
: To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
: For additional commands, e-mail: dev-help@lucene.apache.org
:
:
-Hoss
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Updated] (SOLR-3154) SolrJ CloudServer should be leader and
network aware when adding docs
Posted by "Mark Miller (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-3154?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Mark Miller updated SOLR-3154:
------------------------------
Fix Version/s: 4.1
> SolrJ CloudServer should be leader and network aware when adding docs
> ---------------------------------------------------------------------
>
> Key: SOLR-3154
> URL: https://issues.apache.org/jira/browse/SOLR-3154
> Project: Solr
> Issue Type: Improvement
> Components: SolrCloud
> Affects Versions: 4.0-ALPHA
> Reporter: Grant Ingersoll
> Assignee: Mark Miller
> Priority: Minor
> Fix For: 4.1, 5.0
>
> Attachments: SOLR-3154.patch
>
>
> It would be good when indexing if the SolrJ CloudServer was leader aware so that we could avoid doing an extra hop for the data. It would also be good if one could easily set things up based on data locality principles. This might mean that CloudServer is aware of where on the network it is and would pick leaders that are as close as possible (i.e. local, perhaps.) This would come in to play when working with tools like Hadoop or other grid computing frameworks.
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (SOLR-3154) SolrJ CloudServer should be leader
and network aware when adding docs
Posted by "Mark Miller (Commented) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-3154?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13222359#comment-13222359 ]
Mark Miller commented on SOLR-3154:
-----------------------------------
The main difficulty here is that to get the readable to indexed id that we hash, we need the field type and the id field. The id field we can simply have the user supply - but technically each field type can impl a custom method to get the indexed id that we hash on. Not sure what we do about this, but initially I was thinking we could support one or two field types (int, string) and just copy the (simple) logic for these two types into the server impl. Then the user would also have to supply the type (int,string) of the id field as well - though we could default it to what the example schema uses.
> SolrJ CloudServer should be leader and network aware when adding docs
> ---------------------------------------------------------------------
>
> Key: SOLR-3154
> URL: https://issues.apache.org/jira/browse/SOLR-3154
> Project: Solr
> Issue Type: Improvement
> Components: SolrCloud
> Affects Versions: 4.0
> Reporter: Grant Ingersoll
> Priority: Minor
> Fix For: 4.0
>
>
> It would be good when indexing if the SolrJ CloudServer was leader aware so that we could avoid doing an extra hop for the data. It would also be good if one could easily set things up based on data locality principles. This might mean that CloudServer is aware of where on the network it is and would pick leaders that are as close as possible (i.e. local, perhaps.) This would come in to play when working with tools like Hadoop or other grid computing frameworks.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Commented] (SOLR-3154) SolrJ CloudServer should be leader
and network aware when adding docs
Posted by "Michael Garski (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-3154?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13399612#comment-13399612 ]
Michael Garski commented on SOLR-3154:
--------------------------------------
Would there be any issues with just hashing on the string representation of the ID? That would provide a neutral format that both client and server could use and I implemented that approach in a patch for SOLR-2592 (pluggable sharding), however I have not yet done anything on the SolrJ side of things.
> SolrJ CloudServer should be leader and network aware when adding docs
> ---------------------------------------------------------------------
>
> Key: SOLR-3154
> URL: https://issues.apache.org/jira/browse/SOLR-3154
> Project: Solr
> Issue Type: Improvement
> Components: SolrCloud
> Affects Versions: 4.0
> Reporter: Grant Ingersoll
> Priority: Minor
> Fix For: 4.1
>
>
> It would be good when indexing if the SolrJ CloudServer was leader aware so that we could avoid doing an extra hop for the data. It would also be good if one could easily set things up based on data locality principles. This might mean that CloudServer is aware of where on the network it is and would pick leaders that are as close as possible (i.e. local, perhaps.) This would come in to play when working with tools like Hadoop or other grid computing frameworks.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] [Updated] (SOLR-3154) SolrJ CloudServer should be leader and
network aware when adding docs
Posted by "Grant Ingersoll (Updated) (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-3154?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Grant Ingersoll updated SOLR-3154:
----------------------------------
Description: It would be good when indexing if the SolrJ CloudServer was leader aware so that we could avoid doing an extra hop for the data. It would also be good if one could easily set things up based on data locality principles. This might mean that CloudServer is aware of where on the network it is and would pick leaders that are as close as possible (i.e. local, perhaps.) This would come in to play when working with tools like Hadoop or other grid computing frameworks. (was: It would be good when indexing if the SolrJ CloudServer was leader aware so that we could avoid doing an extra hop for the data.)
Summary: SolrJ CloudServer should be leader and network aware when adding docs (was: SolrJ CloudServer should be leader aware when adding docs)
> SolrJ CloudServer should be leader and network aware when adding docs
> ---------------------------------------------------------------------
>
> Key: SOLR-3154
> URL: https://issues.apache.org/jira/browse/SOLR-3154
> Project: Solr
> Issue Type: Improvement
> Components: SolrCloud
> Affects Versions: 4.0
> Reporter: Grant Ingersoll
> Priority: Minor
> Fix For: 4.0
>
>
> It would be good when indexing if the SolrJ CloudServer was leader aware so that we could avoid doing an extra hop for the data. It would also be good if one could easily set things up based on data locality principles. This might mean that CloudServer is aware of where on the network it is and would pick leaders that are as close as possible (i.e. local, perhaps.) This would come in to play when working with tools like Hadoop or other grid computing frameworks.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org