Showing results for 
Search instead for 
Did you mean: 

performance fine tuning for MSDP

now that i have a new and bigger MSDP drive, i find that the backups speed are still pathetic. i only get 18MB/s for this particular backup test.

the maximum fragment size is 51 MB. that is the ONLY settings configurable from the GUI.

is there another way to fine tune the backup transfer rate?

11/27/2017 17:18:25 - Info dbclient (pid=1536) Backup started
11/27/2017 17:18:25 - Info bpbrm (pid=5234) bptm pid: 5251
11/27/2017 17:18:25 - Info bptm (pid=5251) start
11/27/2017 17:18:25 - connected; connect time: 0:00:00
11/27/2017 17:18:26 - Info bptm (pid=5251) using 1048576 data buffer size
11/27/2017 17:18:26 - Info bptm (pid=5251) setting receive network buffer to 262144 bytes
11/27/2017 17:18:26 - Info bptm (pid=5251) using 256 data buffers
11/27/2017 17:18:26 - Info bptm (pid=5251) start backup
11/27/2017 17:18:26 - Info bptm (pid=5251) backup child process is pid 5255
11/27/2017 17:18:26 - Info dbclient (pid=1536) dbclient(pid=1536) wrote first buffer(size=262144)
11/27/2017 17:18:26 - begin writing



Tags (2)
1 Reply

Re: performance fine tuning for MSDP

This is true for 7.7 as well as beyond:

For the first case when many replication jobs are running at once, but few appear to be making progress, the system may have reached a state of "gridlock."  The solution is simply to decrease the Max I/O streams on the disk pool which is doing the replications.  By default, this number is -1 which indicates unlimited replication jobs can be run on this storage server.  In the NetBackup Administration Console, the Disk Pool properties dialog has a check box to enable limiting the I/O streams (figure):

To increase bandwidth utilization, it may be beneficial to increase the number of replication jobs running concurrently.  This is done by tuning the SLP parameters.  In NetBackup 7.5, the tuning parameters for SLP replication jobs are the same as those used for duplication jobs.  To increase the number of concurrent jobs, the size of the replication batches must be made smaller.  This will force NetBackup to run more simultaneous jobs. 

With the I/O value de-selected (as pictured above), the default setting is 100 streams. Netbackup recommends adjusting per the storage server and enviormental variables but to begin at 80 and work downward or upward depending on how the server processes jobs after the adjustment. The theory is a small adjustment should be made and then monitored. Adjust again and monitor. This should be done and reviewed consistently with the MSDP performance (transfer rate, bffer, dedup ratio, etc.). I would start with the I/O paremeter before adjust the .cfg files. A .cfg file manipulation can result in a temporary fix or workaround where the I/O monitoring will have a long term affect and is easily adjustable.


Tags (1)