TDWI Articles

How to Implement Containers and Virtual Machines on the Same Bare Metal Cloud for Performance and Cost-Efficiency

How can your IT organization enjoy the benefits of virtual machines and containers? MetalSoft's Bernie Wu offers his perspective.

Enterprises interested in great application efficiency and more consistent operations are increasingly turning to containers. In fact, 75 percent of IT leaders expect their use of containers to increase in the next year. The ability to deploy applications both on premises and in the cloud makes containers an ideal solution for working within a bare metal cloud.

For Further Reading:

Executive Q&A: Containers and Kubernetes Accelerating Digital Transformation in 2021

Survey Reveals Cloud Use, Challenges, and Plans

Cloud Deployments: Benefiting User, Developer, and Vendor Communities

On the other hand, enterprises expect that many legacy applications will remain running on virtual machines indefinitely. How can IT organizations see the benefits of both virtual machines and containers?

We talked with Bernie Wu, a datacenter infrastructure software expert at MetalSoft, to learn how a bare metal cloud can be used to host both types of applications while addressing their storage persistency, scaling, performance, and automation considerations in a cost-effective manner.

Upside: Why would companies want to implement containers and virtual machines?

Bernie Wu: Most organizations are currently increasing their use of containers as they adopt cloud native applications and microservices architectures. Applications running in containers also benefit from increased portability, autoscaling, less overhead, and higher efficiencies.

That doesn't mean virtual machines don't have their place in modern infrastructures. Beyond the fact that many legacy applications are difficult to move off virtual machines, VMs also provide more hardened security boundaries and are better suited for more static applications. IT organizations ideally need to be able to implement both containers and virtual workloads on a common, cloud-like infrastructure whether on premises or with a cloud service provider while minimizing the additional management complexity and potential performance and scaling tradeoffs.

What challenges do enterprises face when implementing containers?

Because of costs or security considerations, some enterprises may choose to implement their own container infrastructure either on premises or at a colocation provider. They then face a choice of whether to run their container on bare metal or within a virtual machine environment. If they desire to run their infrastructure on bare metal for maximum performance, they should consider using a comprehensive bare-metal automation platform that can orchestrate capabilities such as cloud-like services, persistent storage, dynamic server scaling, automated provisioning/deployment, isolated L2 networking, and inventory and life cycle management of the hardware infrastructure.

Another challenge of dealing with containers running inside of VMs (instead of directly on bare metal) is the differences in life cycle management of each as well as the differences in approach to network connectivity. Because of the intertwining of these different computing approaches, it will take two levels of administrators to coordinate closely with each other to have the infrastructure operate optimally for both workloads. This adds to the management complexity.

How does a bare-metal cloud address the challenges of implementing containers and virtual machines?

Building an IT infrastructure on a bare metal foundation can boost automation and help facilitate digital transformation by allowing the infrastructure to be managed and consumed as if using a self-service virtual private cloud. Virtual machine and container clusters can be placed in parallel, bare-metal isolated zones that can be separately self-administered.

Features that an enterprise should consider when looking for a bare metal platform include the ability to automatically provision switches to deliver native Layer 2 networking support at the user level to reduce latency and improve performance. The ability to provision both local and network-based persistent storage to both VM and containers is also important. Finally, the ability to automatically provision and deploy both VM and container clusters in a self-service fashion is desirable.

About the Author

James E. Powell is the editorial director of TDWI, including research reports, the Business Intelligence Journal, and Upside newsletter. You can contact him via email here.


TDWI Membership

Accelerate Your Projects,
and Your Career

TDWI Members have access to exclusive research reports, publications, communities and training.

Individual, Student, and Team memberships available.