Have you opened a ticket with HP support? They have the expertise to help you troubleshoot your situation and determine what the expected performance of copy-to-tape is and whether there is a problem.

In our testing we used multiple servers and multiple streams to test performance from servers to VTL. We did not test copy to tape performance as it was out of scope for this report.

We observed about 58MB/sec from a single server, running a single stream to a single Virtual tape on a single node. Our tests were run with data that was uniformly compressible at 2:1.

Your report of 14 hours to copy 1.5TB to tape from VTL represents approximately 31.2 MB/sec. Depending on the type and capacity of the tapes you are transferring to, this may be expected performance, you’ll need to confirm that with HP.

Tony.

In our testing, we did not explore implementation of Data Protector. The ease of deployment section talked about our experience dropping a VLS VTL into an existing Data Protector environment that was already set up and using tape.

Not knowing anything specific about your environment, I can’t comment on your experience, but I’d be happy to chat with you about our testing and methodology and learn more about what you’re trying to do. You can find my contact info here: http://www.enterprisestrategygroup.com/tony-palmer/

Thanks for your interest,
Tony.