Struct CuQuantumConfig 

pub struct CuQuantumConfig {
    pub device_id: i32,
    pub multi_gpu: bool,
    pub num_gpus: usize,
    pub memory_pool_size: usize,
    pub async_execution: bool,
    pub memory_optimization: bool,
    pub precision: ComputePrecision,
    pub gate_fusion_level: GateFusionLevel,
    pub enable_profiling: bool,
    pub max_statevec_qubits: usize,
    pub tensor_contraction: TensorContractionAlgorithm,
    pub enable_tf32: bool,
}

cuQuantum simulation configuration

Fields

device_id: i32

Device ID to use (-1 for auto-select)

multi_gpu: bool

Enable multi-GPU execution

num_gpus: usize

Number of GPUs to use (0 for all available)

memory_pool_size: usize

Memory pool size in bytes (0 for auto)

async_execution: bool

Enable asynchronous execution

memory_optimization: bool

Enable memory optimization (may reduce peak memory)

precision: ComputePrecision

Computation precision

gate_fusion_level: GateFusionLevel

Gate fusion level

enable_profiling: bool

Enable profiling

max_statevec_qubits: usize

Maximum number of qubits for state vector simulation

tensor_contraction: TensorContractionAlgorithm

Tensor network contraction algorithm

enable_tf32: bool

Enable TF32 tensor core mode (NVIDIA Ampere and newer) When enabled, FP32 matrix operations use 19-bit TensorFloat-32 format providing near-FP32 accuracy with ~8x speedup on tensor cores Only effective when device has tensor cores (compute capability ≥ 8.0)

Implementations

impl CuQuantumConfig

pub fn large_circuit() -> Self

Create configuration optimized for large circuits

pub fn variational() -> Self

Create configuration optimized for variational algorithms (VQE/QAOA)

pub fn multi_gpu(num_gpus: usize) -> Self

Create configuration for multi-GPU execution

pub fn with_tf32(self, enable: bool) -> Self

Create configuration with TF32 explicitly enabled/disabled

pub fn should_use_tf32(&self, device_info: &CudaDeviceInfo) -> bool

Check if TF32 should be used based on device capabilities

Trait Implementations

impl Clone for CuQuantumConfig

fn clone(&self) -> CuQuantumConfig

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for CuQuantumConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for CuQuantumConfig

fn default() -> Self

Returns the “default value” for a type.