Safely Experimenting in the Corporate Network with z/VM

With contributions from Alan Altmark, IBM Senior Managing z/VM Consultant

The influential CP-67 computer system, one of the first widely adopted mainframe virtualization solutions and a forerunner to the modern systems in use today, was first released 50 years ago. It was developed by researchers who were looking to explore a “time-sharing computer system,” and decades later, virtualization solutions still empower businesses to experiment and learn from their networks.

“Virtualization enables playtime,” said Alan Altmark, Senior Managing z/VM and Linux IT consultant with IBM Lab Services. “It decreases the cost and risk of experimentation.”

Altmark has led a series of SHARE presentations on the z/VM hypervisor and virtual switches, which enable scalable, secure, and highly available corporate networks.  

In a recent interview, he explained that virtualization also offers a safe way for corporations to support experimentation resourcing work, or the type of projects that help companies rapidly develop, test and deploy application changes or implement system engineering changes.

In a physical-only environment, these types of projects can get expensive. You’d need to take a production machine and separate it into physical partitions, reserving memory and computing resources for each new experiment. But computing resources come at a premium.

“If someone says I need 256 gigabytes for an engineering project, that’s a lot,” he explained. “In z/VM, I can share that space among many virtual servers, and I don’t need to dedicate that much CPU power to a single instance. With z/VM, I can vet changes before I go through the expense of deploying them out into real partitions.”

This helps in a few areas of the business, including network diagnostics. With virtualization, administrators gain what Altmark calls a “virtual data center,” a no-risk environment to explore the inner workings of your network and troubleshoot issues.

“It’s like being able to crawl around under the floor and through the ceilings very easily, with easy access to the infrastructure,” he said. “Or putting your car up on a jack and tinkering with it.”

For example, an administrator may want to take two Linux servers and connect them to a private virtual switch to run traces on Wireshark, a network analysis tool that allows users to examine network traffic.

“Maybe I know how to do it on a Windows laptop but I want to try things on my Linux mainframe. I can do that with a virtual switch,” Altmark said. “With z/VM, I don’t have to run the traffic over my physical network. I can do network experimentation, see what traffic flow looks like, and learn what I need to know for diagnostic purposes and learning purposes.”

These insights can help IT teams make smarter decisions about how they deploy physical resources. It’s part of the reason some of the world’s biggest corporations rely on virtualization to make better use of their large mainframe environments.

“Financial services companies, insurers, government agencies, ecommerce businesses – they’re all running on mainframes,” Altmark said. “z/VM is there to help them move faster and develop new solutions faster.”

That speaks to the heart of why businesses choose virtualization in the first place: the technology helps them achieve more with less.

Alan recently hosted a presentation on the subject of z/VM and virtual switches. To learn more, watch the full video of his session,
z/VM Virtual Switch: The Basics on the SHARE YouTube channel.

Recent Stories
Transforming Batch Resiliency through Greater Control Using Analytics and Automation

Rolling with the Tide: Flexibility in Mainframe Training

Message from SHARE: A Full-Throated Defense of COBOL