Mainframe data: hybrid cloud object store vs. tape

Five years ago, many predicted that the mainframe would soon disappear. But that hasn’t happened. In fact, the number of mainframe workloads is growing, especially now that manufacturers have engineered blades that are ideal for running generative AI. In 2019, half of enterprises surveyed said their number of mainframe workloads had grown; in 2023, 62% said the same.

But while mainframes have advanced, most organizations are still storing their mainframe data in tape or virtual tape libraries (VTL). Stakeholders need mainframe data to be safe, secure, and accessible — and storing data in these archaic environments accomplishes none of these goals.  Meanwhile, enterprises are rapidly moving away from tape and other on-premises storage in favor of cloud object stores. These object stores are stable, resilient, easily accessible, and simpler to manage. Plus, data can be stored as immutable copies which are impervious to ransomware attacks.

For starters, data is not easily accessible on tapes, which are also inherently difficult to protect from disaster and are complex and costly to maintain. Plus, as older employees retire, organizations lose the expertise to manage these systems. Newer, younger employees have little to no experience with tape or VTL, and training is a costly, time-consuming task — assuming there’s anyone left in the organization who can conduct the training.

Any organization that currently stores its mainframe data on tape or VTL should prioritize migrating that data to a hybrid cloud object store. The advantages of hybrid cloud are just too compelling, especially when compared to the inherent risks of the status quo.

An object store in a hybrid cloud environment provides:

• Easy, ubiquitous availability: Those who need access to mainframe data can securely connect to it from anywhere, any time, and with almost any device.
• Cost optimization: Tape-based infrastructure and VTL have heavy capital and operational costs for storage space, maintenance, and hardware. Moving mainframe data to an object store in a hybrid cloud environment enables organizations to pay only for the resources they need at any given time without any capital investment for storage. Plus, it’s simpler to manage, and enables IT to leverage lower-cost storage tiers for archive and other less frequently accessed data, both of which reduces costs. And when an organization needs to scale up or down, it’s simple to do so on a dime.
• Simplification of the environment: Legacy storage systems are complex and often siloed. Object stores in hybrid cloud enable organizations to centralize management of storage without having to deal with storage islands or the complexity of VTL grids.
• Stronger security: Hyperscale cloud providers have invested billions of dollars building a secure infrastructure, and while customers do bear a portion of responsibility for the security of their data, it’s far simpler to achieve a unified, consistent security posture there. Immutable copies render data impervious to ransomware, and data is encrypted by default, which makes it more difficult to steal.
• Resilience: Hyperscale cloud storage is replicated multiple times throughout the infrastructure, and hybrid cloud environments have many excellent additional backup options. In particular, backups stored as immutable copies can’t be compromised in a ransomware attack.
• Workforce: Hybrid cloud environments are simpler to manage and, while cloud experts are in high demand, it’s a growing field, unlike the dwindling pool of tape and VTL experts.
• AI, analytics, and other cloud services: With mainframe data in the cloud, it’s possible to extract a great deal more value by connecting it to AI, analytics, and other powerful services.