f353e00172
- Restructure .env.example with security-focused documentation, service-specific environment file references, and AWS Secrets Manager integration - Update CLI tests workflow to single Python 3.13 version, add pytest-mock dependency, and consolidate test execution with coverage - Add comprehensive security validation to package publishing workflow with manual approval gates, secret scanning, and release
126 lines
3.6 KiB
Python
126 lines
3.6 KiB
Python
"""
|
|
GPU Acceleration Module
|
|
|
|
This module provides a clean, backend-agnostic interface for GPU acceleration
|
|
in the AITBC project. It automatically selects the best available backend
|
|
(CUDA, Apple Silicon, CPU) and provides unified ZK operations.
|
|
|
|
Usage:
|
|
from gpu_acceleration import GPUAccelerationManager, create_gpu_manager
|
|
|
|
# Auto-detect and initialize
|
|
with GPUAccelerationContext() as gpu:
|
|
result = gpu.field_add(a, b)
|
|
metrics = gpu.get_performance_metrics()
|
|
|
|
# Or specify backend
|
|
gpu = create_gpu_manager(backend="cuda")
|
|
result = gpu.field_mul(a, b)
|
|
"""
|
|
|
|
# Public API
|
|
from .gpu_manager import (
|
|
GPUAccelerationManager,
|
|
GPUAccelerationContext,
|
|
create_gpu_manager,
|
|
get_available_backends,
|
|
auto_detect_best_backend,
|
|
ZKOperationConfig
|
|
)
|
|
|
|
# Backend enumeration
|
|
from .compute_provider import ComputeBackend, ComputeDevice
|
|
|
|
# Version information
|
|
__version__ = "1.0.0"
|
|
__author__ = "AITBC Team"
|
|
__email__ = "dev@aitbc.dev"
|
|
|
|
# Initialize logging
|
|
import logging
|
|
logger = logging.getLogger(__name__)
|
|
|
|
# Auto-detect available backends on import
|
|
try:
|
|
AVAILABLE_BACKENDS = get_available_backends()
|
|
BEST_BACKEND = auto_detect_best_backend()
|
|
logger.info(f"GPU Acceleration Module loaded")
|
|
logger.info(f"Available backends: {AVAILABLE_BACKENDS}")
|
|
logger.info(f"Best backend: {BEST_BACKEND}")
|
|
except Exception as e:
|
|
logger.warning(f"GPU backend auto-detection failed: {e}")
|
|
AVAILABLE_BACKENDS = ["cpu"]
|
|
BEST_BACKEND = "cpu"
|
|
|
|
# Convenience functions for quick usage
|
|
def quick_field_add(a, b, backend=None):
|
|
"""Quick field addition with auto-initialization."""
|
|
with GPUAccelerationContext(backend=backend) as gpu:
|
|
return gpu.field_add(a, b)
|
|
|
|
def quick_field_mul(a, b, backend=None):
|
|
"""Quick field multiplication with auto-initialization."""
|
|
with GPUAccelerationContext(backend=backend) as gpu:
|
|
return gpu.field_mul(a, b)
|
|
|
|
def quick_field_inverse(a, backend=None):
|
|
"""Quick field inversion with auto-initialization."""
|
|
with GPUAccelerationContext(backend=backend) as gpu:
|
|
return gpu.field_inverse(a)
|
|
|
|
def quick_multi_scalar_mul(scalars, points, backend=None):
|
|
"""Quick multi-scalar multiplication with auto-initialization."""
|
|
with GPUAccelerationContext(backend=backend) as gpu:
|
|
return gpu.multi_scalar_mul(scalars, points)
|
|
|
|
# Export all public components
|
|
__all__ = [
|
|
# Main classes
|
|
"GPUAccelerationManager",
|
|
"GPUAccelerationContext",
|
|
|
|
# Factory functions
|
|
"create_gpu_manager",
|
|
"get_available_backends",
|
|
"auto_detect_best_backend",
|
|
|
|
# Configuration
|
|
"ZKOperationConfig",
|
|
"ComputeBackend",
|
|
"ComputeDevice",
|
|
|
|
# Quick functions
|
|
"quick_field_add",
|
|
"quick_field_mul",
|
|
"quick_field_inverse",
|
|
"quick_multi_scalar_mul",
|
|
|
|
# Module info
|
|
"__version__",
|
|
"AVAILABLE_BACKENDS",
|
|
"BEST_BACKEND"
|
|
]
|
|
|
|
# Module initialization check
|
|
def is_available():
|
|
"""Check if GPU acceleration is available."""
|
|
return len(AVAILABLE_BACKENDS) > 0
|
|
|
|
def is_gpu_available():
|
|
"""Check if any GPU backend is available."""
|
|
gpu_backends = ["cuda", "apple_silicon", "rocm", "opencl"]
|
|
return any(backend in AVAILABLE_BACKENDS for backend in gpu_backends)
|
|
|
|
def get_system_info():
|
|
"""Get system information for GPU acceleration."""
|
|
return {
|
|
"version": __version__,
|
|
"available_backends": AVAILABLE_BACKENDS,
|
|
"best_backend": BEST_BACKEND,
|
|
"gpu_available": is_gpu_available(),
|
|
"cpu_available": "cpu" in AVAILABLE_BACKENDS
|
|
}
|
|
|
|
# Initialize module with system info
|
|
logger.info(f"GPU Acceleration System Info: {get_system_info()}")
|