[jira] [Comment Edited] (HDFS-8818) Allow Balancer to run faster
[ https://issues.apache.org/jira/browse/HDFS-8818?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15995525#comment-15995525 ] Tsz Wo Nicholas Sze edited comment on HDFS-8818 at 5/3/17 7:58 PM: --- 15*72 = 1080. I guess you were using default dfs.balancer.moverThreads, which is 1,000. Have you tried to increased it, say 10,000 or 30,000? was (Author: szetszwo): 15*72 = 1080. I guess you were using default fs.balancer.moverThreads, which is 1,000. Have you tried to increased it, say 10,000 or 30,000? > Allow Balancer to run faster > > > Key: HDFS-8818 > URL: https://issues.apache.org/jira/browse/HDFS-8818 > Project: Hadoop HDFS > Issue Type: Sub-task > Components: balancer & mover >Reporter: Tsz Wo Nicholas Sze >Assignee: Tsz Wo Nicholas Sze > Fix For: 2.8.0, 2.7.4, 3.0.0-alpha1 > > Attachments: bal1.png, bal2.png, h8818_20150723.patch, > h8818_20150727.patch, HDFS-8818-branch-2.7.00.patch > > > The original design of Balancer is intentionally to make it run slowly so > that the balancing activities won't affect the normal cluster activities and > the running jobs. > There are new use case that cluster admin may choose to balance the cluster > when the cluster load is low, or in a maintain window. So that we should > have an option to allow Balancer to run faster. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Comment Edited] (HDFS-8818) Allow Balancer to run faster
[ https://issues.apache.org/jira/browse/HDFS-8818?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15995477#comment-15995477 ] Kihwal Lee edited comment on HDFS-8818 at 5/3/17 7:17 PM: -- I set dfs.datanode.balance.max.concurrent.moves to 15, which is what we used to run with 2.7. 71 thread pools with 15 threads in each created. The 72th one is small and then started to fail. So I still see "skipping..." message in the log. The throughput is still visibly lower, after a minute of initial spike. The iteration lasts a bit longer, but the average throughput is still similarly low. Other config values are all set to default. I believe I have presented enough evidence. was (Author: kihwal): I set the dfs.datanode.balance.max.concurrent.moves to 15, which is what we used to use with 2.7. 71 thread pools with 15 threads in each created. The 72th one is small and then started to fail. So I still see "skipping..." message in the log. The throughput is still visibly lower, after a minute of initial spike. The iteration lasts a bit longer, but the average throughput is still similarly low. Other config values are all set to default. I believe I have presented enough evidence. > Allow Balancer to run faster > > > Key: HDFS-8818 > URL: https://issues.apache.org/jira/browse/HDFS-8818 > Project: Hadoop HDFS > Issue Type: Sub-task > Components: balancer & mover >Reporter: Tsz Wo Nicholas Sze >Assignee: Tsz Wo Nicholas Sze > Fix For: 2.8.0, 2.7.4, 3.0.0-alpha1 > > Attachments: bal1.png, bal2.png, h8818_20150723.patch, > h8818_20150727.patch, HDFS-8818-branch-2.7.00.patch > > > The original design of Balancer is intentionally to make it run slowly so > that the balancing activities won't affect the normal cluster activities and > the running jobs. > There are new use case that cluster admin may choose to balance the cluster > when the cluster load is low, or in a maintain window. So that we should > have an option to allow Balancer to run faster. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Comment Edited] (HDFS-8818) Allow Balancer to run faster
[ https://issues.apache.org/jira/browse/HDFS-8818?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15993664#comment-15993664 ] Wei-Chiu Chuang edited comment on HDFS-8818 at 5/2/17 8:25 PM: --- [~kihwal] FWIW, the hang you mentioned was fixed in HDFS-11377. was (Author: jojochuang): [~kihwal] FWIW, the hung you mentioned is fixed in HDFS-11377. > Allow Balancer to run faster > > > Key: HDFS-8818 > URL: https://issues.apache.org/jira/browse/HDFS-8818 > Project: Hadoop HDFS > Issue Type: Sub-task > Components: balancer & mover >Reporter: Tsz Wo Nicholas Sze >Assignee: Tsz Wo Nicholas Sze > Fix For: 2.8.0, 2.7.4, 3.0.0-alpha1 > > Attachments: h8818_20150723.patch, h8818_20150727.patch, > HDFS-8818-branch-2.7.00.patch > > > The original design of Balancer is intentionally to make it run slowly so > that the balancing activities won't affect the normal cluster activities and > the running jobs. > There are new use case that cluster admin may choose to balance the cluster > when the cluster load is low, or in a maintain window. So that we should > have an option to allow Balancer to run faster. -- This message was sent by Atlassian JIRA (v6.3.15#6346) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org