| Feature | Description |
|---------|-------------|
| Core | 8‑core ARM Cortex‑X2, out‑of‑order, 3.2 GHz boost, supporting Scalable Vector Extension (SVE) up to 2048 bit. |
| Cache hierarchy | L1 I/D 64 KB per core, L2 256 KB per core, unified L3 16 MB (inclusive). |
| Security | ARM TrustZone‑v2 + JunYun Secure Enclave (SAE‑3) – hardware‑rooted attestation, encrypted VM support. |
| Instruction set extensions | - MVE‑AI: SIMD ops optimized for matrix multiplication
- CME (Cache‑Matrix Engine) for zero‑copy data streaming. |
| Power management | Fine‑grained DVFS per core + cluster, with hardware‑controlled “sleep‑gate” for sub‑10 mW idle. |
| Application | How the juy‑108 Helps | |-------------|-----------------------| | Wearable Health Tracker | Continuous temperature & humidity monitoring, motion detection for step counting, BLE for smartphone sync. | | Smart Home Climate Sensor | Reports indoor climate to a hub; low‑power mode lets it run on a coin cell for a year. | | Industrial Asset Monitoring | Detects vibration anomalies (via accelerometer) and sends alerts over BLE to a gateway that forwards to a cloud dashboard. | | Educational Kit | Plug‑and‑play with Arduino IDE; students can quickly build projects that combine sensing and wireless communication. | juy-108
| Layer | Tools / SDKs | Highlights |
|-------|--------------|------------|
| OS | Linux‑5.15 (Yocto), Zephyr RTOS (for low‑latency), Windows 11 (via WSL) | Full driver stack, pre‑emptible scheduling for AI kernels. |
| Runtime | J‑Runtime (lightweight), OpenCL‑v3 (experimental) | J‑Runtime exposes Zero‑Copy API (jTensorMap()) and Secure Compute Zones. |
| Compilers | J‑MLIR (based on LLVM‑MLIR), J‑LLVM (for native code), J‑CUDA (CUDA‑compatible). | Auto‑vectorization of SVE, quantization-aware training support. |
| Frameworks | Plugins for TensorFlow 2.x, PyTorch 2.0, ONNX Runtime, MXNet | One‑click conversion scripts (juy_convert.py). |
| Debug/Profiling | J‑Trace (cycle‑accurate trace), Perf‑J (perf‑compatible), J‑Profiler GUI | Real‑time heat‑map of tensor engine utilisation. |
| Security | SAE‑3 SDK (remote attestation, sealed storage) | Enables confidential AI inference for edge‑cloud split. | | Feature | Description | |---------|-------------| | Core
Tip for developers: When targeting the J‑Tensor engine, keep tensor dimensions multiples of 8 (for systolic array alignment) and use BF16 if you need a good balance between precision and throughput. The J‑MLIR optimizer will automatically pad to the next multiple when necessary. | Layer | Tools / SDKs | Highlights