After VMware: what alternatives are there for edge and remote site infrastructure?

Broadcom’s acquisition of VMware in 2023 was accompanied by significant changes to licensing. Perpetual licenses were discontinued, the product portfolio was reorganized into bundles, and for many organizations, costs rose significantly. This has accelerated the search for alternatives, particularly for environments where VMware has historically been widely used: edge locations and remote sites.

Why is the edge a seperate issue?

A central data center environment typically has sufficient space, power, and management capacity for a full-fledged virtualization stack. An edge location, such as a branch office, a factory floor, or a retail location, does not have that luxury. There, you’re dealing with two or three servers, limited space, and, ideally, no local IT administrator to maintain the environment.

VMware was commonly used in those environments, but in many cases, the licensing costs are no longer proportional to the scale of the installation. This makes the search for lighter alternatives a priority.

Scale Computing HC3

Scale Computing provides a hyperconverged infrastructure platform called HC3. Hyperconverged means that compute, storage, and network virtualization are offered in a single integrated stack, running on standard x86 hardware.

HC3 is built on KVM as a hypervisor and uses its own distributed storage system. The platform is designed for easy management: an installation on two or three nodes is up and running within an hour, without requiring in-depth knowledge of virtualization.

Scale Computing specifically targets organizations with many small locations. Centralized management of dozens or hundreds of HC3 installations through a single interface is a core feature of the platform. Updates and configuration changes can be rolled out from that central location.

StorMagic SvSAN

StorMagic SvSAN is a software-defined storage solution that provides high availability across two nodes. That is the minimum number of servers required for a redundant environment—and thus exactly what many edge locations have available.

SvSAN runs as a virtual appliance on the existing hypervisor, whether it’s KVM, VMware, or Hyper-V. It mirrors data between the two nodes and ensures that the environment remains available if one node fails. A quorum mechanism using a lightweight witness prevents split-brain situations.

StorMagic focuses on use cases where high availability is required but a full cluster of three or more nodes is not feasible due to cost or physical constraints.

When do you choose one or the other?

Scale Computing HC3 and StorMagic SvSAN are not direct competitors. They address overlapping but distinct challenges.

  • HC3 is a complete hyperconverged stack. It allows you to replace the entire virtualization layer, including storage and compute.
  • SvSAN is a storage layer that is deployed alongside an existing hypervisor. If the virtualization layer already exists and only the storage needs to be replaced or enhanced, SvSAN is a less invasive solution.

In environments where VMware is being replaced, SvSAN can be combined with an alternative hypervisor. In environments where a complete replacement of the stack is desired, HC3 is an option as a complete replacement.

What does this mean for hardware selection?

Both Scale Computing and StorMagic run on standard x86 hardware. Scale Computing also provides its own certified nodes, built on hardware from Supermicro and other manufacturers. StorMagic is hardware-independent and runs on virtually any x86 server.

This allows for flexibility in hardware selection. Choosing the right servers in terms of CPU, memory, and storage configuration has a direct impact on the platform’s performance.

Lees ook.