[jira] [Commented] (CASSANDRA-15600) Algorithmic token allocation does not handle the racks = RF case well
[ https://issues.apache.org/jira/browse/CASSANDRA-15600?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17136177#comment-17136177 ] Ekaterina Dimitrova commented on CASSANDRA-15600: - CASSANDRA-15877 > Algorithmic token allocation does not handle the racks = RF case well > - > > Key: CASSANDRA-15600 > URL: https://issues.apache.org/jira/browse/CASSANDRA-15600 > Project: Cassandra > Issue Type: Bug > Components: Feature/Virtual Nodes >Reporter: Branimir Lambov >Assignee: Ekaterina Dimitrova >Priority: Normal > Fix For: 4.0, 4.0-alpha4 > > Attachments: CASSANDRA-15600 unit.txt, CASSANDRA-15600-jvm.txt.zip, > Screen Shot 2020-03-02 at 6.55.41 PM.png > > > When the number of racks in a data centre is equal to the replication factor, > the node allocation algorithm may calculate the target ownership incorrectly > and allocate unsuitable tokens. This causes some inefficiency when the number > of nodes in the DC is small, and significant problems if the number of nodes > in each rack is different. > In this case (racks count equal to replication factor) the load within each > rack is effectively distributed independently. The allocation algorithm > should reflect that, restricting the allocation space to the rack (instead of > the DC) and using the RF=1 solution. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-15600) Algorithmic token allocation does not handle the racks = RF case well
[ https://issues.apache.org/jira/browse/CASSANDRA-15600?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17133313#comment-17133313 ] Ekaterina Dimitrova commented on CASSANDRA-15600: - Hi [~kornelpal], I am on it. Thanks! > Algorithmic token allocation does not handle the racks = RF case well > - > > Key: CASSANDRA-15600 > URL: https://issues.apache.org/jira/browse/CASSANDRA-15600 > Project: Cassandra > Issue Type: Bug > Components: Feature/Virtual Nodes >Reporter: Branimir Lambov >Assignee: Ekaterina Dimitrova >Priority: Normal > Fix For: 4.0, 4.0-alpha4 > > Attachments: CASSANDRA-15600 unit.txt, CASSANDRA-15600-jvm.txt.zip, > Screen Shot 2020-03-02 at 6.55.41 PM.png > > > When the number of racks in a data centre is equal to the replication factor, > the node allocation algorithm may calculate the target ownership incorrectly > and allocate unsuitable tokens. This causes some inefficiency when the number > of nodes in the DC is small, and significant problems if the number of nodes > in each rack is different. > In this case (racks count equal to replication factor) the load within each > rack is effectively distributed independently. The allocation algorithm > should reflect that, restricting the allocation space to the rack (instead of > the DC) and using the RF=1 solution. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-15600) Algorithmic token allocation does not handle the racks = RF case well
[ https://issues.apache.org/jira/browse/CASSANDRA-15600?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17121134#comment-17121134 ] Kornel Pal commented on CASSANDRA-15600: [~e.dimitrova] you have added a generateSplits method to ReplicationAwareTokenAllocator, but it does not seem to be used. New comments suggest that splits are supposed to be used, but generateRandomTokens is used instead. Could you please double check whether some intended changes are missing from the commit. > Algorithmic token allocation does not handle the racks = RF case well > - > > Key: CASSANDRA-15600 > URL: https://issues.apache.org/jira/browse/CASSANDRA-15600 > Project: Cassandra > Issue Type: Bug > Components: Feature/Virtual Nodes >Reporter: Branimir Lambov >Assignee: Ekaterina Dimitrova >Priority: Normal > Fix For: 4.0, 4.0-alpha4 > > Attachments: CASSANDRA-15600 unit.txt, CASSANDRA-15600-jvm.txt.zip, > Screen Shot 2020-03-02 at 6.55.41 PM.png > > > When the number of racks in a data centre is equal to the replication factor, > the node allocation algorithm may calculate the target ownership incorrectly > and allocate unsuitable tokens. This causes some inefficiency when the number > of nodes in the DC is small, and significant problems if the number of nodes > in each rack is different. > In this case (racks count equal to replication factor) the load within each > rack is effectively distributed independently. The allocation algorithm > should reflect that, restricting the allocation space to the rack (instead of > the DC) and using the RF=1 solution. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-15600) Algorithmic token allocation does not handle the racks = RF case well
[ https://issues.apache.org/jira/browse/CASSANDRA-15600?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17050341#comment-17050341 ] Brandon Williams commented on CASSANDRA-15600: -- Committed, thanks! > Algorithmic token allocation does not handle the racks = RF case well > - > > Key: CASSANDRA-15600 > URL: https://issues.apache.org/jira/browse/CASSANDRA-15600 > Project: Cassandra > Issue Type: Bug > Components: Feature/Virtual Nodes >Reporter: Branimir Lambov >Assignee: Ekaterina Dimitrova >Priority: Normal > Fix For: 4.0-alpha > > Attachments: CASSANDRA-15600 unit.txt, CASSANDRA-15600-jvm.txt.zip, > Screen Shot 2020-03-02 at 6.55.41 PM.png > > > When the number of racks in a data centre is equal to the replication factor, > the node allocation algorithm may calculate the target ownership incorrectly > and allocate unsuitable tokens. This causes some inefficiency when the number > of nodes in the DC is small, and significant problems if the number of nodes > in each rack is different. > In this case (racks count equal to replication factor) the load within each > rack is effectively distributed independently. The allocation algorithm > should reflect that, restricting the allocation space to the rack (instead of > the DC) and using the RF=1 solution. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-15600) Algorithmic token allocation does not handle the racks = RF case well
[ https://issues.apache.org/jira/browse/CASSANDRA-15600?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17049781#comment-17049781 ] Ekaterina Dimitrova commented on CASSANDRA-15600: - [~brandon.williams] please review > Algorithmic token allocation does not handle the racks = RF case well > - > > Key: CASSANDRA-15600 > URL: https://issues.apache.org/jira/browse/CASSANDRA-15600 > Project: Cassandra > Issue Type: Bug > Components: Feature/Virtual Nodes >Reporter: Branimir Lambov >Assignee: Ekaterina Dimitrova >Priority: Normal > Attachments: CASSANDRA-15600 unit.txt, CASSANDRA-15600-jvm.txt.zip, > Screen Shot 2020-03-02 at 6.55.41 PM.png > > > When the number of racks in a data centre is equal to the replication factor, > the node allocation algorithm may calculate the target ownership incorrectly > and allocate unsuitable tokens. This causes some inefficiency when the number > of nodes in the DC is small, and significant problems if the number of nodes > in each rack is different. > In this case (racks count equal to replication factor) the load within each > rack is effectively distributed independently. The allocation algorithm > should reflect that, restricting the allocation space to the rack (instead of > the DC) and using the RF=1 solution. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-15600) Algorithmic token allocation does not handle the racks = RF case well
[ https://issues.apache.org/jira/browse/CASSANDRA-15600?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17049778#comment-17049778 ] Ekaterina Dimitrova commented on CASSANDRA-15600: - [~lor...@datastax.com] we will need to document this improvement. Please let me know if you need any additional/ticket information. Thanks. > Algorithmic token allocation does not handle the racks = RF case well > - > > Key: CASSANDRA-15600 > URL: https://issues.apache.org/jira/browse/CASSANDRA-15600 > Project: Cassandra > Issue Type: Bug > Components: Feature/Virtual Nodes >Reporter: Branimir Lambov >Assignee: Ekaterina Dimitrova >Priority: Normal > Attachments: CASSANDRA-15600 unit.txt, CASSANDRA-15600-jvm.txt.zip, > Screen Shot 2020-03-02 at 6.55.41 PM.png > > > When the number of racks in a data centre is equal to the replication factor, > the node allocation algorithm may calculate the target ownership incorrectly > and allocate unsuitable tokens. This causes some inefficiency when the number > of nodes in the DC is small, and significant problems if the number of nodes > in each rack is different. > In this case (racks count equal to replication factor) the load within each > rack is effectively distributed independently. The allocation algorithm > should reflect that, restricting the allocation space to the rack (instead of > the DC) and using the RF=1 solution. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-15600) Algorithmic token allocation does not handle the racks = RF case well
[ https://issues.apache.org/jira/browse/CASSANDRA-15600?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17049771#comment-17049771 ] Ekaterina Dimitrova commented on CASSANDRA-15600: - CI results attached. Two of the dtests turned out flakey on trunk, no new issue introduced by this patch. Respective issues were raised - CASSANDRA-15613 and CASSANDRA-15614 Patch: [pull request|https://github.com/ekaterinadimitrova2/cassandra/pull/20] [branch|https://github.com/ekaterinadimitrova2/cassandra/tree/CASSANDRA-15600] > Algorithmic token allocation does not handle the racks = RF case well > - > > Key: CASSANDRA-15600 > URL: https://issues.apache.org/jira/browse/CASSANDRA-15600 > Project: Cassandra > Issue Type: Bug > Components: Feature/Virtual Nodes >Reporter: Branimir Lambov >Assignee: Ekaterina Dimitrova >Priority: Normal > Attachments: CASSANDRA-15600 unit.txt, CASSANDRA-15600-jvm.txt.zip, > Screen Shot 2020-03-02 at 6.55.41 PM.png > > > When the number of racks in a data centre is equal to the replication factor, > the node allocation algorithm may calculate the target ownership incorrectly > and allocate unsuitable tokens. This causes some inefficiency when the number > of nodes in the DC is small, and significant problems if the number of nodes > in each rack is different. > In this case (racks count equal to replication factor) the load within each > rack is effectively distributed independently. The allocation algorithm > should reflect that, restricting the allocation space to the rack (instead of > the DC) and using the RF=1 solution. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org