Metalvisor, as a TypeZero hypervisor, Has no orchestration layer at the virtual machine (VM) layer and has no oversubscription. This is very different from Type 1 & 2 hypervisors.
Remove Virtualization Overhead / Tax
The virtualization tax/overhead is removed, and workloads can utilize close to 100% of the hardware.
Metalvisor does not share hardware between VMs. This gives guest VMs the same profile as bare-metal and also how we got the name Metalvisor.
Quality of Service
Metalvisor Isolates & Dedicates Hardware to each VM; Cores, Cache, Threads, Memory, PCIe, Network, Storage
No Noisy Neighbors
Metalvisor removes 100% of the Noisy Neighbor issues that present on traditional virtualization that cause poor performance and QoS.
Workloads have execution cycles the same as the previous one, resulting in dependable & interference free compute environment.
Configurable Cache Ways
Workloads can dial up or down the levels of cache ways assigned to each VM resulting in configurable QoS.
Metalvisor is based on Red Hat Enterprise Linux and can be deployed on bare metal from many different OEMs. Metalvisor is launched from firmware/UEFI layer and is as close to the metal as possible.
The Orchestration Domain (OD) is a RHEL VM instance used to work with LibVirt to start/stop VMs and other VM related functions. The OD also uses Cockpit (open source web interface) for managing workloads.
Active Response Capability (ARC)
Metalvisor uses a bare metal application to enforce policy across VMs, known as active response capability (ARC). ARC ensures things like secure boot, cryptography, software exploit, and hardware protection are enforced.
Oops! Something went wrong while submitting the form.