Offering Microsecond-Scale Cross-VM Core Elasticity on Colocated Lightweight Virtual Machines

2026-08-12

AI Summary

This work addresses the challenge of achieving microsecond-scale dynamic physical core reallocation in high-density, lightweight virtual machine (VM) environments to simultaneously optimize tail latency and resource utilization—capabilities lacking in existing solutions. To this end, the authors propose HyperFlux, a KVM-based ultra-lightweight virtualization substrate that enables, for the first time, cross-VM microsecond-level core migration and elastic parallelism scaling. Key innovations include runtime dynamic core redistribution, a forced core reclamation mechanism, and a low-overhead scheduler. Experimental results demonstrate that HyperFlux achieves core migration in just 13 μs, with per-VM memory footprint of 3.2 MB and cold-start latency of 1.37 ms. Under high load, it reduces tail latency for high-priority VMs by up to an order of magnitude compared to Firecracker and Cloud Hypervisor.

Abstract

Serverless platforms commonly colocate many diverse workloads, each in a fast-booting, memory-lean virtual machine (VM), to improve deployment density. Overprovisioning each VM for its peak protects tail latency during traffic bursts but hurts density; maintaining high density while effectively protecting tail latency requires the infrastructure to be able to shift physical cores, at a microsecond timescale, to whichever latency-sensitive VM is bursting and reclaim them as the burst subsides. No VM substrate delivers this: conventional VMs resize a guest's cores only through a millisecond-scale vCPU hot-plug path, Firecracker fixes a VM's core count at boot, and the ultralight VMs that boot fastest drop multicore execution entirely. We present HyperFlux, a commodity-KVM ultralight VM substrate that makes a VM's parallelism width (the number of physical cores backing it) elastic at runtime. We show that HyperFlux can move a core across VMs in merely 13$μ$s, even when forcibly reclaiming it from a busy donor, orders of magnitude faster than vCPU hot-plug. A HyperFlux VM incurs only a 3.2MB memory footprint and can cold-boot in 1.37ms, on par with the fastest-booting ultralight VMs, while uniquely supporting multicore parallelism. Under colocation, it can reduce high-priority VMs' tail latency by up to 10x under high load compared to static core-sharing with Firecracker and Cloud Hypervisor, and deliver a lower and more stable tail latency compared to using cgroup and vCPU hot-plug under changing load bursts.

Problem

• core elasticity
• tail latency
• colocated VMs
• serverless platforms
• resource overprovisioning

Innovation

• microsecond-scale core elasticity
• ultralight VM
• cross-VM resource reallocation
• tail latency optimization
• runtime parallelism scaling