[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jeremiah Jordan updated CASSANDRA-5220: --- Fix Version/s: 3.0 beta 1 Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Marcus Olsson Labels: performance, repair Fix For: 3.0 beta 1 Attachments: 5220-yourkit.png, 5220-yourkit.tar.bz2, cassandra-3.0-5220-1.patch, cassandra-3.0-5220-2.patch, cassandra-3.0-5220.patch Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jonathan Ellis updated CASSANDRA-5220: -- Assignee: Marcus Olsson Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Marcus Olsson Labels: performance, repair Attachments: 5220-yourkit.png, 5220-yourkit.tar.bz2, cassandra-3.0-5220-1.patch, cassandra-3.0-5220-2.patch, cassandra-3.0-5220.patch Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jonathan Ellis updated CASSANDRA-5220: -- Reviewer: Stefania (was: Yuki Morishita) Reassigning review to [~Stefania] Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Marcus Olsson Labels: performance, repair Attachments: 5220-yourkit.png, 5220-yourkit.tar.bz2, cassandra-3.0-5220-1.patch, cassandra-3.0-5220-2.patch, cassandra-3.0-5220.patch Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marcus Olsson updated CASSANDRA-5220: - Attachment: cassandra-3.0-5220-2.patch With this patch the dtest works for me, both with vnodes and without. There was a problem with wrap around ranges and the merkle tree iterator I made that only occurred if there was data around both the min and max tokens. The problem was that it had already iterated over the merkle tree for the wrap around range in the beginning, so the solution I chose was to reiterate over the merkle trees in Validator.add() if the end was reached. With the patch applied I saw a ~4 seconds improvement(from ~18 to ~14) when using vnodes. Without vnodes it took around 10-11 seconds. Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Labels: performance, repair Attachments: 5220-yourkit.png, 5220-yourkit.tar.bz2, cassandra-3.0-5220-1.patch, cassandra-3.0-5220-2.patch, cassandra-3.0-5220.patch Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marcus Olsson updated CASSANDRA-5220: - Attachment: cassandra-3.0-5220-1.patch This is the correct patch, which is based on the latest trunk. It's created with diff instead of format-patch, hope that it won't cause any problems. Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Labels: performance, repair Attachments: 5220-yourkit.png, 5220-yourkit.tar.bz2, cassandra-3.0-5220-1.patch, cassandra-3.0-5220.patch Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Marcus Olsson updated CASSANDRA-5220: - Attachment: cassandra-3.0-5220.patch Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Yuki Morishita Labels: performance, repair Attachments: 5220-yourkit.png, 5220-yourkit.tar.bz2, cassandra-3.0-5220.patch Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yuki Morishita updated CASSANDRA-5220: -- Fix Version/s: (was: 2.1.1) 3.0 Re-targeting this to 3.0 since my current work involves messaging change to handle multiple ranges at once, and is based on major refactoring being done in CASSANDRA-6455. Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Yuki Morishita Labels: performance, repair Fix For: 3.0 Attachments: 5220-yourkit.png, 5220-yourkit.tar.bz2 Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jonathan Ellis updated CASSANDRA-5220: -- Fix Version/s: (was: 2.1 rc1) 2.1.1 Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Yuki Morishita Labels: performance, repair Fix For: 2.1.1 Attachments: 5220-yourkit.png, 5220-yourkit.tar.bz2 Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ryan McGuire updated CASSANDRA-5220: Attachment: 5220-yourkit.png I don't have a lot of experience with java profiling, but this is a yourkit comparison between the two runs of the dtest. Old Time is without vnodes, New Time is with vnodes. I turned off internode_compression because that run showed all the time being spent in Snappy compressor code, but even without compression the repair time didn't improve (much): !5220-yourkit.png! Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Yuki Morishita Labels: performance, repair Fix For: 2.1 beta2 Attachments: 5220-yourkit.png Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ryan McGuire updated CASSANDRA-5220: Attachment: 5220-yourkit.tar.bz2 I attached 5220-yourkit.tar.bz2 which is snapshots made in yourkit of various runs of the dtest. Maybe someone with more experience with yourkit can glean some more info out of it. Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Yuki Morishita Labels: performance, repair Fix For: 2.1 beta2 Attachments: 5220-yourkit.png, 5220-yourkit.tar.bz2 Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Benedict updated CASSANDRA-5220: Labels: performance repair (was: ) Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Yuki Morishita Labels: performance, repair Fix For: 2.1 beta2 Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jonathan Ellis updated CASSANDRA-5220: -- Fix Version/s: (was: 2.0) 2.1 Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Yuki Morishita Fix For: 2.1 Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- 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
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jonathan Ellis updated CASSANDRA-5220: -- Fix Version/s: (was: 1.2.4) 2.0 Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Yuki Morishita Fix For: 2.0 Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- 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
[jira] [Updated] (CASSANDRA-5220) Repair improvements when using vnodes
[ https://issues.apache.org/jira/browse/CASSANDRA-5220?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sylvain Lebresne updated CASSANDRA-5220: Fix Version/s: (was: 1.2.2) 1.2.3 Repair improvements when using vnodes - Key: CASSANDRA-5220 URL: https://issues.apache.org/jira/browse/CASSANDRA-5220 Project: Cassandra Issue Type: Improvement Components: Core Affects Versions: 1.2.0 beta 1 Reporter: Brandon Williams Assignee: Yuki Morishita Fix For: 1.2.3 Currently when using vnodes, repair takes much longer to complete than without them. This appears at least in part because it's using a session per range and processing them sequentially. This generates a lot of log spam with vnodes, and while being gentler and lighter on hard disk deployments, ssd-based deployments would often prefer that repair be as fast as possible. -- 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