Add GPU fan control, live telemetry, and per-profile fan curves in HyperSwap dashboard

server.py

@@ -11,6 +11,7 @@ from pydantic import BaseModel, Field
 
 import ram_optimizer
 import vram_arbitrator
+import overclock_manager
 
 logging.basicConfig(level=logging.INFO, format="%(asctime)s [%(levelname)s] %(name)s: %(message)s")
 logger = logging.getLogger("model_manager_server")
@@ -31,6 +32,14 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
+@app.on_event("startup")
+async def on_startup():
+    await vram_arbitrator.arbitrator.start()
+
+@app.on_event("shutdown")
+async def on_shutdown():
+    await vram_arbitrator.arbitrator.stop()
+
 # Pydantic Request Models
 class SwitchRequest(BaseModel):
     model: str = Field(..., description="Name of the Ollama model to hot-swap to in VRAM", example="qwen3.8fast:latest")
@@ -44,6 +53,17 @@ class BenchmarkRequest(BaseModel):
     iterations: Optional[int] = Field(2, description="Number of back-and-forth switch iterations to measure", example=2)
     models: Optional[List[str]] = Field(None, description="Optional pair of models to benchmark between", example=["qwen3.8fast:latest", "smtek/Qwen3.8-27B:Q2_K_XL"])
 
+class OverclockApplyRequest(BaseModel):
+    profile: str = Field(..., description="Profile name: ollama | comfy | balanced", example="ollama")
+
+class OverclockProfileUpdate(BaseModel):
+    config: Dict[str, Any] = Field(..., description="Profile settings dict", example={"power_limit_w": 370, "core_offset_mhz": 100})
+
+class FanRequest(BaseModel):
+    mode: str = Field("auto", description="'auto' or 'manual'", example="manual")
+    percent: Optional[int] = Field(None, description="Fan speed 30-100 when mode=manual", example=70)
+    speed_pct: Optional[int] = Field(None, description="Alias for percent (30-100)", example=70)
+
 
 # ==========================================
 # REST API ENDPOINTS
@@ -58,6 +78,7 @@ async def get_all_stats() -> Dict[str, Any]:
     comfy_state = await vram_arbitrator.get_comfyui_live_state()
     history = vram_arbitrator.get_switch_history()
     comfy_models = ram_optimizer.find_comfy_model_files()
+    arbitrator_status = vram_arbitrator.arbitrator.get_status()
     
     return {
         "timestamp": asyncio.get_event_loop().time(),
@@ -65,10 +86,12 @@ async def get_all_stats() -> Dict[str, Any]:
         "ram": mem_stats,
         "ollama": ollama_state,
         "comfyui": comfy_state,
+        "arbitrator": arbitrator_status,
         "history": history,
         "comfy_models_count": len(comfy_models),
     }
 
+
 @app.get("/api/gpu", summary="GPU Sensors and VRAM Breakdown", tags=["Telemetry"])
 async def get_gpu_metrics() -> Dict[str, Any]:
     """Retrieve detailed NVML sensors (utilization %, temp, power, fan, clocks, and per-process VRAM allocation)."""
@@ -160,6 +183,51 @@ async def api_run_benchmark(req: BenchmarkRequest):
     res_str = await run_model_switch_benchmark(iterations=req.iterations or 2)
     return json.loads(res_str)
 
+# ==========================================
+# OVERCLOCK MANAGEMENT
+# ==========================================
+
+@app.get("/api/overclock", summary="Overclock Status & Profiles", tags=["Overclock"])
+async def api_overclock_status():
+    """Get live GPU overclock state, active profile, and all per-app profiles."""
+    return overclock_manager.get_status()
+
+@app.post("/api/overclock/apply", summary="Apply Overclock Profile", tags=["Overclock"])
+async def api_overclock_apply(req: OverclockApplyRequest):
+    """Apply a named overclock profile (ollama | comfy | balanced) to the GPU immediately."""
+    res = overclock_manager.apply_profile(req.profile)
+    if not res.get("success"):
+        raise HTTPException(status_code=400, detail=res.get("error"))
+    return res
+
+@app.get("/api/overclock/profiles", summary="List Overclock Profiles", tags=["Overclock"])
+async def api_overclock_profiles():
+    """List all overclock profiles with their current settings."""
+    return overclock_manager.get_profiles()
+
+@app.post("/api/overclock/profiles/{name}", summary="Update Overclock Profile", tags=["Overclock"])
+async def api_overclock_update_profile(name: str, req: OverclockProfileUpdate):
+    """Update a profile's settings (persisted to disk)."""
+    res = overclock_manager.set_profile(name, req.config)
+    if not res.get("success"):
+        raise HTTPException(status_code=400, detail=res.get("error"))
+    return {"success": True, "profile": name, "profiles": res.get("profiles")}
+
+@app.get("/api/overclock/fan", summary="Get GPU Fan Status", tags=["Overclock"])
+@app.get("/api/gpu/fan", summary="Get GPU Fan Status", tags=["Overclock"])
+async def api_get_fan_status():
+    """Get current GPU fan control mode and speed."""
+    return overclock_manager.get_fan_status()
+
+@app.post("/api/overclock/fan", summary="Set GPU Fan Speed", tags=["Overclock"])
+@app.post("/api/gpu/fan", summary="Set GPU Fan Speed", tags=["Overclock"])
+async def api_set_fan(req: FanRequest):
+    """Set the GPU fan to manual speed (30-100%) or back to automatic control."""
+    pct = req.percent if req.percent is not None else req.speed_pct
+    if req.mode == "manual" and pct is not None:
+        return overclock_manager.set_fan_speed(pct)
+    return overclock_manager.set_fan_auto()
+
 # Mount static web UI files
 app.mount("/static", StaticFiles(directory="/home/drjones/unified-model-manager/static"), name="static")