Mainframe Disaster Recovery without Tapes – Part 2

Bookmark and Share
Dave Race

Dave Race

Traditionally, we relied on tapes for mainframe disaster recovery, but today we have neither the time nor the patience to wait for linear tapes. Recovery time objectives (RTO), which used to be measured in days, are now measured in hours. Recovery point objectives (RPO) have shrunk from days to minutes. There is a plethora of solutions for our open system servers and storage but for the mainframe our choices are somewhat limited.

The combination of virtual tape with data de-dupe technology is revolutionizing mainframe disaster recovery. We can reliably integrate FICON channel gateways (CG) emulating 3590 tape mechanisms with data de-duplicating virtual tape libraries, enabling us to recover mainframe environments in a fraction of the time it used to take to recover from tapes. This is particularly significant for shops that rely on Sun Guard for their disaster recovery solution.

I have a customer who recently conducted a DR exercise at Sun Guard and were given only 24 hours to complete it. Their previous DR exercise recovered from tape and took about 16 hours just to restore their data before they could begin testing their systems. Needless to say, their results were less than satisfactory. With this new CG and de-dupe technology, we were able to recovery all of their data in just a couple of hours and they were able to successfully complete all of their system tests ahead of schedule! Happiness is…

The set-up is pretty simple at a high level. FICON was laid between the customer’s Z9 and the channel gateways (CG). The CG emulated 3590 tape mechanism and wrote to a data de-duplicating virtual tape library (VTL). That VTL replicated to another smaller VTL which had a CG attached to it as well. The Sun Guard mainframe read the customer’s data through the second set of channel gateways and presto – very fast recovery. Now there were a few wrinkles in the program that made this exercise particularly exciting, but we’ll get into that next time…

This is part 2 of this series of posts. Read Part 1 and Part 3.

Dave Race

Solutions Architect


Tags: , , ,

2 Responses to “Mainframe Disaster Recovery without Tapes – Part 2”

  1. […] Disaster Recovery without Tapes I covered “virtual tape libraries”. In part 2, I covered “the disaster recovery process”.  In this post I will address “additional challenges associated with mainframe disaster […]

  2. […] is all great, but how does this help disaster recovery? I’ll go into that in Part 2, and Part […]

Contact Us Request a Consultation