Cristalink Support Forums

As an update, we have identified what seems to be causing the issue in this situation. Our SOFS shares were created with the Continuous Availability option. Firestreamer was writing with IO sizes of 32 K directly to the disk, making it slower. When we created shares to the same location without the ...

We access using \\[clustername]\share. All our tapes are setup this way. Copying to [clustername] is fast.

I have tried to copy a 7 GB ISO file to the network locations. Copying to the SSD pool: Max speed seen is 130 MB/s Copying to the normal disk pool: Max Speed seen is about 110 MB/s Much faster than any operation to backup to tape. So I know I can send data quickly to the SOFS pools. I don't know why...

I tried all the settings for the registry keys, and none of them changed the speed at which the backups were taken.

Is it normal that I had to create the keys, or were they supposed to be there in the first place?

Good afternoon.

We will try to change the registry keys. I understand we do not have to reboot the server, but we will need to restart the Firestreamer service, correct?

Thank you

We have DPM 2012 R2 Update Rollup 4 with Firestreamer 4.0 virtual tape libraries, writing to tapes on a SOFS share. We access the shares over multiple 10g links. There is one share (named 3) that has some of the hard disks on it that are SSDs that are used as write caches, and one share (named 4) th...