The company’s own site names even more targeted use-cases beyond those three, ranging from high core count ARM servers to RISC-V systems built for one kernel per cluster on heterogeneous SoCs.Sandbox is marketed for running AI agents in their own kernel with direct GPU access, and LiveUpdate promises to patch a running kernel without taking the system down.This release only supports x86_64 hardware, though Multikernel Technologies says the architecture-specific code is already modular enough for other ports to follow later.
How it works
The host tracks all of this through device tree overlays under /sys/fs/multikernel/, and it can shift hardware between running instances while everything stays up.
It’s FOSS has been helping people use Linux for the past 14 years. Help us stay independent from big tech. Become a Plus member, enjoy ad-free reading and get 5 eBooks.
And here’s the source code for your perusal. 👇Multikernel avoids both problems. Each instance is a full, independent kernel running on its own dedicated cores. Modern NICs and SSDs can hand out separate hardware queues too, so an instance gets exclusive access to a slice of the device, not just a slice of the CPU.This app-kernel and device-kernel split is the pattern the company documents for the architecture generally, dedicating one instance purely to drivers and I/O so the app-facing kernels never see a device interrupt at all.
Who’s this for?
There’s more to it than what I have covered so far. You can read the release announcement to dive into the comprehensive benchmark numbers shared by Cong.One kernel runs as the host, controlling the machine’s CPUs, memory, and PCI devices. When it spins up a new kernel instance, that instance gets its own dedicated slice of hardware and boots straight onto it using Linux’s kexec mechanism, running natively rather than under a hypervisor.Multikernel Technologies has put out mklinux v7.0-mk2, the first public release of the multikernel Linux tree it’s been building since last year.Pay once, Enjoy foreverDon’t worry, it isn’t some obscure copy but rather a direct fork of the Linux kernel with a bunch of patches added that give it such a high level of parallel play.Multikernel Technologies already sells three products based on this kernel tree, and what they’re pitched as says plenty about who this is actually for. Private Cloud is sold as a way to consolidate workloads on bare metal without a hypervisor. OEMs looking to bake Multikernel into an embedded device are also welcome.The basic premise of the project is that one physical server can run several independent Linux kernels side by side, each with its own dedicated CPUs and memory. So, the same box could run a database on one kernel and a GPU training job on another without either kernel touching the other’s resources.
Every one of those pitches targets infrastructure and SRE teams running their own hardware, not desktop users.
That’s a different setup from both VMs and containers. A VM still puts a hypervisor and a host kernel between the app and the hardware. A container skips the hypervisor, but every tenant on the machine shares one kernel.
Go lifetime
