Interface SIMDBackend

All Superinterfaces:

Backend, ComputeBackend

All Known Implementing Classes:

AbstractNativeSIMDLinearAlgebraBackend, EpistemeLinearAlgebraBackend, NativeSIMDComplexBackend, NativeSIMDRealBackend

public interface SIMDBackend
extends ComputeBackend

Interface for SIMD-accelerated compute backends.

Implementations use the JDK Vector API (jdk.incubator.vector) or native SIMD intrinsics via FFM/Panama to accelerate numerical operations.

Example implementations:

• SIMDRealLinearAlgebraProvider — vectorized linear algebra
• Future: NEON (ARM), SVE, AVX-512 specialized backends

Since:

1.2

Author:

Silvere Martin-Michiellot, Gemini AI (Google DeepMind)

Method Summary

Modifier and Type	Method	Description
default HardwareAccelerator	getAcceleratorType()	Returns the type of hardware accelerator used by this backend.
int	getPreferredVectorWidth()	Returns the preferred vector width in bytes for double operations.
String	getSimdLevel()	Returns the SIMD instruction set level available.
default String	getType()	Returns the backend type category (e.g., "plotting", "molecular", "tensor", "math").

Methods inherited from interface Backend

getAlgorithmProviders, getName, getPriority, getStatusMessage, isAvailable, isExplicitlyDisabled, shutdown

Methods inherited from interface ComputeBackend

createBackend, createContext, getDescription, getId, supportsComplexNumbers, supportsFloatingPoint, supportsParallelOps

Method Details

getType

default String getType()

Description copied from interface: Backend

Returns the backend type category (e.g., "plotting", "molecular", "tensor", "math").

Specified by:

getType in interface Backend

Specified by:

getType in interface ComputeBackend

getAcceleratorType

default HardwareAccelerator getAcceleratorType()

Description copied from interface: ComputeBackend

Returns the type of hardware accelerator used by this backend.

Specified by:

getAcceleratorType in interface ComputeBackend

Returns:

the accelerator type

getSimdLevel

String getSimdLevel()

Returns the SIMD instruction set level available.

Examples: "AVX2", "AVX-512", "NEON", "SVE", "GENERIC"

getPreferredVectorWidth

int getPreferredVectorWidth()

Returns the preferred vector width in bytes for double operations.

Typical values: 32 (AVX2), 64 (AVX-512), 16 (NEON)