synos-hive-profiler — Arcanum Hive Hardware Profiler
tags: [hive mesh distributed]
synos-hive-profiler — Arcanum Hive Hardware Profiler
Section titled “synos-hive-profiler — Arcanum Hive Hardware Profiler”Classification: PUBLIC
Crate: fruit/crates/synos-hive-profiler/
Milestone: v79+
License: MIT OR Apache-2.0
What It Is
Section titled “What It Is”synos-hive-profiler is the Arcanum Hive hardware profiler. It detects node
capabilities and classifies tiers, enabling the hive controller to schedule
work where it makes sense.
Architecture
Section titled “Architecture”Capability Detection
Section titled “Capability Detection”| Capability | Detection Method |
|---|---|
| CPU cores | /proc/cpuinfo parsing |
| Memory | /proc/meminfo |
| GPU | PCI device enumeration + driver detection |
| NVMe | Block device detection |
| Network | Interface enumeration + speed detection |
| TPM | TPM2 device detection |
| Secure boot | EFI variable inspection |
Tier Classification
Section titled “Tier Classification”| Tier | Description |
|---|---|
tier-0 | High-end workstation (32+ cores, 128GB+ RAM, GPU) |
tier-1 | Mid-range workstation (8-31 cores, 32-127GB RAM) |
tier-2 | Low-end device (4-7 cores, 8-31GB RAM) |
tier-3 | Embedded/IoT (1-3 cores, <8GB RAM) |
How It’s Wired
Section titled “How It’s Wired”- synos-hive-controller — receives profiler results for workload scheduling
- synos-hive-bft — consensus on node capability assessments
- synos-ops — displays profiler results in the 23-tab dashboard
- synos-audit-trail — logs all profiling events
Future Ideas
Section titled “Future Ideas”- Dynamic profiling — re-profile nodes periodically for capacity changes
- Specialization detection — identify GPU/NVMe/TPM specializations
- Health scoring — combine capability with reliability metrics
- Cost modeling — estimate energy/performance ratio for scheduling