Re: [Gluster-users] glusterfs replica volume self heal dir very slow!!why?
Hi Paul Robert Marino and Roman thank you very much for you advise! to Paul Robert Marino: In fact, I use XFS as the underlying filesystem in my test cases。 NOW, According to your advise, I will increse the background self heal count to test, I will do a feedback after my test, in addition,Do you know change XFS which options can improve lots of small file performance?。 thank you very much!I'm looking forward to your reply! justgluste...@gmail.com 发件人: Paul Robert Marino 发送时间: 2014-09-08 00:15 收件人: justgluste...@gmail.com; gluster-devel 抄送: gluster-users 主题: Re: [Gluster-users] glusterfs replica volume self heal dir very slow!!why? Its the small files there is an overhead on any operation on small lots of small file this is not unique to Gluster. Also are you using XFS As the underlying filesystem if you are not that would play a big part ext has issues with performance when dealing with small file due to its over reliance on inodes for read operations. I would advise you to experiment with incrementing the background self heal count a little at a time till you achieve optimal performance. Unfortunately I can't give you hard numbers because this kind of tuning is more of an art than a science because of the number of possible variables which come into play depending on your hardware configuration and exact use case. -- Sent from my HP Pre3 On Sep 5, 2014 4:16 AM, justgluste...@gmail.com wrote: Hi all: I do the following test: I create a glusterfs replica volume (replica count is 2 ) with two server node(server A and server B), then mount the volume in client node, then, I shut down the network of server A node, in client node, I copy a dir(which has a lot of small files), the dir size is 2.9GByte, when copy finish, I start the network of server A node, now, glusterfs self-heal-daemon start heal dir from server B to server A, in the end, I find the self-heal-daemon heal the dir use 40 minutes, It's too slow! why? I find out related options with self-heal, as follow: cluster.self-heal-window-size cluster.self-heal-readdir-size cluster.background-self-heal-count I want to ask, modify the above options can improve the performance of heal dir? if possible, please give a reasonable value about above options。 thanks! justgluste...@gmail.com ___ Gluster-users mailing list Gluster-users@gluster.org http://supercolony.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] glusterfs replica volume self heal dir very slow!!why?
Its the small files there is an overhead on any operation on small lots of small file this is not unique to Gluster. Also are you using XFS As the underlying filesystem if you are not that would play a big part ext has issues with performance when dealing with small file due to its over reliance on inodes for read operations.I would advise you to experiment with incrementing the background self heal count a little at a time till you achieve optimal performance. Unfortunately I can't give you hard numbers because this kind of tuning is more of an art than a science because of the number of possible variables which come into play depending on your hardware configuration and exact use case.-- Sent from my HP Pre3On Sep 5, 2014 4:16 AM, justgluste...@gmail.com wrote: Hi all: I do the following test: I create a glusterfs replica volume (replica count is 2 ) with two server node(server A and server B), then mount the volume in client node, then, I shut down the network of server A node, in client node, I copy a dir(which has a lot of small files), the dir size is 2.9GByte, when copy finish, I start the network of server A node, now, glusterfs self-heal-daemon start heal dir from server B to server A, in the end, I find the self-heal-daemon heal the dir use 40 minutes, It's too slow! why? I find out related options with self-heal, as follow: cluster.self-heal-window-size cluster.self-heal-readdir-size cluster.background-self-heal-count I want to ask, modify the above options can improve the performance of heal dir? if possible, please give a reasonable value about above options。 thanks! justgluste...@gmail.com ___ Gluster-users mailing list Gluster-users@gluster.org http://supercolony.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] glusterfs replica volume self heal dir very slow!!why?
Hi, consider looking at this topic http://supercolony.gluster.org/pipermail/gluster-users/2014-September/018577.html . This guy made lots of test. It seems like 50 MB/s (or around 400 Mbps) is the theoretical maximal limit for replica 1. I am waiting for comments to this too. 2014-09-05 11:15 GMT+03:00 justgluste...@gmail.com : > Hi all: > I do the following test: > I create a glusterfs replica volume (replica count is 2 ) with two > server node(server A and server B), then mount the volume in client node, > then, I shut down the network of server A node, in client node, I > copy a dir(which has a lot of small files), the dir size is *2.9GByte,* > when copy finish, I start the network of server A node, now, > glusterfs self-heal-daemon start heal dir from server B to server A, > in the end, I find the self-heal-daemon heal the dir *use 40 > m**inutes, > * *It's too slow! why?* > >I find out related options with self-heal, as follow: > * cluster.self-heal-window-size* > * cluster.self-heal-readdir-size* > * cluster.background-self-heal-count* > > I want to ask, modify the above options can improve the > performance of heal dir? if possible, please give a reasonable value > about above options。 > > thanks! > -- > justgluste...@gmail.com > > ___ > Gluster-users mailing list > Gluster-users@gluster.org > http://supercolony.gluster.org/mailman/listinfo/gluster-users > -- Best regards, Roman. ___ Gluster-users mailing list Gluster-users@gluster.org http://supercolony.gluster.org/mailman/listinfo/gluster-users