Is Your Mainframe Disaster Recovery Strategy Ready for 2020?

In an era of cloud computing, distributed infrastructures and smart devices, many IT departments still rely on legacy disaster recovery (DR) solutions for IBM systems. During an outage with these legacy DR approaches, organizations could lose 24 to 48 hours’ worth of data and business. While this may have been tolerable years ago, it’s no longer acceptable in today’s world.

“They’re as obsolete as the road map in your glove compartment,” asserted Orry Dubois, CTO of ZPX Cloud, a TierPoint partner that specializes in data recovery as a service (DRaaS).

Companies that do business online can rarely afford more than a few minutes of downtime or lost data,  noted Dubois, who spoke on the need to modernize IBM system DR strategies in the webcast Not Your Grandpa’s DR: Delivering Modern Data Availability for IBM Systems.

Legacy disaster recovery technology shows its cracks

Disaster recovery technology has evolved significantly since the time of nightly tape backups and days-long recovery periods. Unfortunately, many organizations are still leveraging DR solutions for their IBM systems based on those old methods.

“Companies who depend on them are at a complete disadvantage,” he said, noting that the typical backup is 24 hours apart. “The amount of data they lose by restoring to the last backup is totally unacceptable.”

These legacy DR environments may be either hosted at a third-party facility with backup and storage services, or they may be in-house systems. Both scenarios are prone to data loss and long recovery windows.

Third-party backup services can protect data in case of a physical disaster, as the data is stored off-site, but they can’t usually provide a quick recovery or enable frequent testing of the DR service. Infrequent testing often means the backup and recovery doesn’t work when its needed. In addition, the recovery process may be manual, dependent on the client to execute it, and time-consuming.

In-house DR offers more control over the equipment, processes and testing schedules. However, it also requires investments in hardware, software and facility maintenance. Moreover, in-house IT staff may lack DR expertise and not trust their own recovery process.

“I’ve seen companies who were totally down with an outage, but [the IT staff] struggled for hours or even days trying to fix the problem instead of invoking their DR program. Because they knew it may not work as planned,” he commented.

Another issue is the all-or-nothing nature of legacy IBM systems DR, said Dubois. Most organizations today operate in multi-cloud or hybrid IT environments which legacy DR wasn’t designed to protect, said Dubois. In these decentralized environments, data and workloads are distributed across multiple cloud providers or hybrid environments. An outage may occur in just one or two locations, not the entire system, making a traditional all-or-nothing backup and restore process challenging.

Businesses look to modern DR technology as the answer

For a business to survive a natural disaster, power outage or cyber-attack, it needs a modern DR solution that can restore operations within minutes and recover data to within an hour or even seconds from the outage.

Dubois suggests taking an “always-on” approach to DR. That means continuous replication of data, automated failover procedures and constant on-demand testing capabilities.

“It’s not something you should have to schedule. You should be able to do on-demand testing, with 24×7 availability,” he advised. “Think about how to automate workflows to make things seamless. Automate the recovery processes at the component level, instead of an all-or-nothing recovery.”

He also recommends evolving your IBM systems DR strategy to focus on data and to develop metrics for it that look at data integrity, availability and confidentiality in the DR process. While recovery times are obviously critical, so is the accuracy, availability and security of the data. With such large volumes of data traveling across the internet and corporate networks, recovery systems must be rooted in data protection best practices.

“We need to shift our thinking about DR from hardware-centric to data-centric solutions,” he said. “We also need to talk about DR as business resumption not recovery. Because the digitization of just about everything has made businesses completely dependent on IT to function.”

Watch the full disaster recovery webinar on infrastructure solutions and ZPX Cloud

Along with TierPoint’s director of product management, PJ Farmer, Dubois also detailed the infrastructure of modern, cloud-based DR solutions for IBM systems and discussed how ZPX Cloud and TierPoint can provide a flexible, fast, and reliable DRaaS solution. Watch the entire webcast Not Your Grandpa’s DR: Delivering Modern Data Availability for IBM Systems.