Re: gluster performance

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi,
Try iperf in TCP mode, because GlusterFS use TCP not UDP.

Best,
Bart

On 25 Oct 2018, at 09:52, Krishna Verma <kverma@xxxxxxxxxxx> wrote:

Hi All,

 

I have checked the network speed bandwidth from my Gluster Master node to Slave node that I found  957 Mbits/sec. But syncing over the geo replication is too much slow as it took 1.28hrs to sync a 9.8GB of file from master to slave.

 

Anyone having any clue to improve the sync speed over geo replication?

 

[root@XXXXXX]# iperf -c sj-gluster01 -u -b 1500m

------------------------------------------------------------

Client connecting to sj-gluster01, UDP port 5001

Sending 1470 byte datagrams, IPG target: 7.84 us (kalman adjust)

UDP buffer size:  208 KByte (default)

------------------------------------------------------------

[  3] local 172.23.16.38 port 39104 connected with 172.18.33.52 port 5001

[ ID] Interval       Transfer     Bandwidth

[  3]  0.0-10.0 sec  1.11 GBytes   957 Mbits/sec

[  3] Sent 813830 datagrams

[  3] Server Report:

[  3]  0.0-10.2 sec   230 MBytes   188 Mbits/sec  12.987 ms 649728/813829 (80%)

[root@XXXXXX]#

 

/Krishna

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
https://lists.gluster.org/mailman/listinfo/gluster-users
_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
https://lists.gluster.org/mailman/listinfo/gluster-users

[Index of Archives]     [Gluster Development]     [Linux Filesytems Development]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux