370f6653e2
The model file is inert. The runtime and its compute target are the phenomenon. This crate is how the runtime KNOWS its compute — the corrective to a CPU-only substrate in a family that built fabric everywhere on purpose. DETECT — live probes only: CPU/NEON (32 cores, bit-exact reference oracle), AMX via a real cblas_sgemm probe, Metal GPU via MPS device query (M3 Ultra). REACH — what ambient credentials open, nothing contacted: CUDA fleets (Nebius, Lambda, Thunder, AWS, GCP, Azure, OVH, Scaleway, Linode), Groq LPU, Cerebras WSE, SambaNova RDU, Tenstorrent via Koyeb. Reachable never serves without a live executor handshake. PLACE — cheapest-sufficient applied to silicon: projected uJ per op from per-member costs (Measured/Benchmarked/Estimated provenance, never self-upgraded) + per-class dispatch floors + the measured m>=8 matrix-block crossover. Recurrences stay on CPU; batched matmuls go to the matrix block. HONESTY GATES (tested): Uncertified members cannot serve — the Metal GPU is detected but refuses work until parity-certified against the CPU oracle (next wave, then the BF16 forwards ask the fabric instead of hard-coding NEON). Efficiency by abstention is DOA. Live on this box (examples/discover): local/cpu-neon Detected BitExact serving local/amx Detected Tolerance(128ulp) serving local/metal-gpu Detected Uncertified gated Also: backend-apple gains free-function sgemm_f32 and mps_device_name/mps_matmul_fp32 re-exports. 11 tests. Costs are v0 projections; IOReport calibration lands with the loader integration. Co-Authored-By: Claude Fable 5 <[email protected]>
41 lines
1.6 KiB
Rust
41 lines
1.6 KiB
Rust
//! Print the compute fabric as seen from this process:
|
|
//! detected local silicon + credential-reachable remote
|
|
//! members, then a few sample placements.
|
|
//!
|
|
//! ```bash
|
|
//! cargo run --release -p jouleclaw-fabric --example discover
|
|
//! # with the infrastructure env loaded:
|
|
//! set -a; source path/to/.env; set +a
|
|
//! cargo run --release -p jouleclaw-fabric --example discover
|
|
//! ```
|
|
|
|
use jouleclaw_fabric::{place, Fabric, OpClass, OpDesc};
|
|
|
|
fn main() {
|
|
let fabric = Fabric::discover();
|
|
println!("compute fabric, as seen from this process:\n");
|
|
for m in &fabric.members {
|
|
println!(
|
|
" {:<16} {:?}/{:?} presence={:?} parity={:?}",
|
|
m.id, m.class, m.locality, m.presence, m.parity,
|
|
);
|
|
println!(" {:<16} {}", "", m.detail);
|
|
}
|
|
let serving: Vec<&str> = fabric.serving().map(|m| m.id.as_str()).collect();
|
|
println!("\nserving now (detected + local + certified): {serving:?}\n");
|
|
|
|
println!("sample placements:");
|
|
for (label, op) in [
|
|
("decode expert matmul (m=1, 4 MFLOP)",
|
|
OpDesc { class: OpClass::MatMul, flops: 4e6, m: 1 }),
|
|
("prefill projection (m=32, 1 GFLOP)",
|
|
OpDesc { class: OpClass::MatMul, flops: 1e9, m: 32 }),
|
|
("lm_head (m=1, 450 MFLOP)",
|
|
OpDesc { class: OpClass::MatMul, flops: 4.5e8, m: 1 }),
|
|
("SSD recurrence (1 GFLOP, sequential)",
|
|
OpDesc { class: OpClass::Recurrence, flops: 1e9, m: 1 }),
|
|
] {
|
|
let p = place(&fabric, &op);
|
|
println!(" {label:<44} → {}", p.because);
|
|
}
|
|
}
|