Replace hotplug cpu by online/offline CPU

[mikhail-sakhnov]

Problem description / Motivation

According to qemu documentation, hot plugging is not supported on arm. Marking cpu as online/offline is supported. We need to change acpi scripts used on x86_64 for hot plugging to support online/offline cpus on arm.

https://docs.kernel.org/next/arch/arm64/cpu-hotplug.html

https://www.qemu.org/docs/master/system/cpu-hotplug.html

Feature idea(s) / DoD

Implementation ideas

[stradig]

I suggest to move that to the start and also switch to online/offline on x86. I believe that will autoscaling also more robust on x86.

[mikhail-sakhnov]

CPU Scaling Feature via online/offline cpu state

Current Implementation

In the existing NeonVM platform, CPU hotplugging is used to dynamically scale CPU resources up or down. The NeonVM controller (neonvm/controllers/vm_controller.go) sends commands via QMP to the QEMU hypervisor. When CPUs are added or removed, a udev rule (neonvm/tools/vm-builder/files/vminit) inside the VM automatically marks the newly added CPUs as online.

While this method works effectively on x86_64 architectures, the ARM architecture in QEMU does not support CPU hotplugging.

Proposed Solution for ARM Architecture

To support CPU scaling on ARM, we will implement a different approach. Instead of hotplugging, all potential CPUs will be allocated to the VM instance during startup. Any CPUs beyond the current scaling limit will be marked as offline. The control of CPU states (online/offline) will no longer rely on the udev rule but will be handled through a VM control daemon running inside the VM.

Key Aspects:

• Initial CPU Allocation: All possible CPUs will be assigned to the VM at startup. Only the CPUs required within the scaling limit will be marked as online, while the others remain offline.
• Daemon for CPU State Management: The introduction of a VM control daemon will enable direct control over the online/offline status of the CPUs. This daemon will expose an API to manage the number of active CPUs.

Required Changes

• NeonVM Controller Modification: Update the NeonVM controller to communicate with the VM control daemon API via vm-runner API for CPU scaling actions instead of using QMP directly to the QEMU monitor.
• VM Runner API: For enhanced security, the daemon will be exposed through the VM Runner API to ensure secure interaction.
• VM Control Daemon Implementation: Develop the VM control daemon that will run inside the VM to manage CPU states.
• Expose VM Control Daemon: Integrate the control daemon via the VM Runner interface. Initial implementation is defined in the PR https://github.com/neondatabase/autoscaling/pull/1090
• VM Builder Update: Modify the VM Builder to include the VM control daemon as part of the VM image, ensuring it is present and operational in all deployed instances.

[mikhail-sakhnov]

both related PRs are in review: #1104 and #1111

1090 been merged already.