Batched storage
The io_uring completion path keeps storage moving efficiently while vCPU limits remain enforced.
Performance
Faster by design.
FlashVMM outperforms Firecracker on every measured cold-boot path, delivering 5.4× to 29.2× faster startup across x86_64, aarch64, bare metal, and nested virtualization. Snapshot restore reaches 1.36 ms on AWS Graviton3 and remains in the low single-digit milliseconds across x86 environments. These results measure the moment the guest is ready and running workload code—the performance that matters in production.
Cold boot that changes the economics of compute
FlashVMM starts faster in every environment shown below. From local x86_64 hardware to AWS Graviton3, the advantage is consistent and decisive.
| Environment | FlashVMM | Firecracker | Performance advantage |
|---|---|---|---|
| Local x86_64 bare metal | 72.6 ms | 396.1 ms | FlashVMM ~5.5× faster |
| Raspberry Pi 5 (aarch64) | 8.65 ms | 176.9 ms | FlashVMM ~20.4× faster |
| AWS Graviton3 bare metal | 4.58 ms | 133.7 ms | FlashVMM ~29.2× faster |
| AWS nested virt (x86_64) | 126.9 ms | 679.2 ms | FlashVMM ~5.4× faster |
Single-digit millisecond startup on aarch64. A greater than 5× advantage on x86_64, including nested virtualization. FlashVMM makes fast, isolated compute available across the fleet.
Restore in milliseconds
A FlashVMM snapshot returns a workload to active execution without paying the cost of another boot. On x86_64 and AWS Graviton3, FlashVMM restores 2.7× to 6.6× faster than Firecracker.
| Environment | FlashVMM | Firecracker | Result |
|---|---|---|---|
| Local x86_64 bare metal | 2.67 ms | 8.9 ms | FlashVMM ~3.3× faster |
| Raspberry Pi 5 (aarch64) | 20.73 ms | 15.1 ms | Native GICv2 restore |
| AWS Graviton3 bare metal | 1.36 ms | 9.0 ms | FlashVMM ~6.6× faster |
| AWS nested virt (x86_64) | 3.71 ms | 10.1 ms | FlashVMM ~2.7× faster |
FlashVMM supports snapshot and restore across x86_64 and aarch64, including both GICv2 and GICv3. The same lifecycle spans local hardware, cloud bare metal, edge systems, and nested virtualization.
Performance that holds under pressure
Sustained reliability
A 1,500-request reliability run completed successfully from start to finish.
Security without startup drag
Capability removal is synchronous and completes before guest execution, preserving a predictable startup path.
Portable performance
One architecture delivers fast boot and restore across x86_64, aarch64, bare metal, and nested virtualization.