Class NativeCUDASparseLinearAlgebraDoubleBackend<E extends FieldElement<E>>

java.lang.Object

org.episteme.nativ.mathematics.linearalgebra.backends.NativeCUDASparseLinearAlgebraDoubleBackend<E>

All Implemented Interfaces:

AutoCloseable, LinearAlgebraProvider<E>, SparseLinearAlgebraProvider<E>, AlgorithmProvider, Backend, ComputeBackend, GPUBackend, NativeBackend

@AutoService({Backend.class,ComputeBackend.class,NativeBackend.class,LinearAlgebraProvider.class,SparseLinearAlgebraProvider.class,GPUBackend.class})
public class NativeCUDASparseLinearAlgebraDoubleBackend<E extends FieldElement<E>>
extends Object
implements SparseLinearAlgebraProvider<E>, NativeBackend, GPUBackend

Native CUDA Sparse Linear Algebra Backend for Double precision.

Nested Class Summary

Nested classes/interfaces inherited from interface GPUBackend

GPUBackend.DeviceInfo

Constructor Summary

Constructors

Constructor	Description
NativeCUDASparseLinearAlgebraDoubleBackend()

Method Summary

Modifier and Type	Method	Description
Vector<E>	add(Vector<E> a, Vector<E> b)
long	allocateGPUMemory(long sizeBytes)	Allocates GPU memory and returns a handle.
Vector<E>	bicgstab(Matrix<E> a, Vector<E> b, Vector<E> x0, E tolerance, int maxIterations)	Solves Ax = b using BiCGSTAB (BiConjugate Gradient Stabilized) method.
void	close()
Vector<E>	conjugateGradient(Matrix<E> a, Vector<E> b, Vector<E> x0, E tolerance, int maxIterations)	Solves Ax = b using Conjugate Gradient method.
Matrix<E>	conjugateTranspose(Matrix<E> a)
void	copyFromGPU(long gpuHandle, DoubleBuffer hostBuffer, long sizeBytes)	Copies data from GPU to host.
void	copyToGPU(long gpuHandle, DoubleBuffer hostBuffer, long sizeBytes)	Copies data from host to GPU.
ExecutionContext	createContext()	Creates an execution context for running operations.
E	dot(Vector<E> a, Vector<E> b)
void	freeGPUMemory(long gpuHandle)	Frees GPU memory.
Matrix<E>	fromDense(Matrix<E> a)	Converts a dense matrix to its sparse representation.
HardwareAccelerator	getAcceleratorType()	Returns the type of hardware accelerator used by this backend.
GPUBackend.DeviceInfo[]	getDevices()	Returns information about available GPU devices.
String	getId()	Returns the unique identifier for this backend (e.g., "javafx", "jmol", "cuda").
String	getName()	Returns the display name for UI presentation.
String	getNativeLibraryName()	Returns the name of the native library this backend depends on.
int	getPriority()	Priority of this provider (higher means more preferred).
String	getType()	Returns the backend type category (e.g., "plotting", "molecular", "tensor", "math").
boolean	isAvailable()	Checks if the provider is available in the current environment.
boolean	isCompatible(Ring<?> ring)	Checks if this provider is compatible with the given ring.
boolean	isExplicitlyDisabled()	Checks if this backend is specifically disabled via a system property: "episteme.backend.disable." + getId().
boolean	isLoaded()	Checks if the native library is loaded and available.
void	matrixMultiply(DoubleBuffer A, DoubleBuffer B, DoubleBuffer C, int m, int n, int k)	Performs matrix multiplication C = A * B on the GPU.
Vector<E>	multiply(Matrix<E> a, Vector<E> x)
E	norm(Vector<E> a)
void	selectDevice(int deviceId)	Selects a GPU device for computation.
void	shutdown()	Called when the provider is no longer needed (e.g., application shutdown).
Vector<E>	solve(Matrix<E> a, Vector<E> b)
Vector<E>	subtract(Vector<E> a, Vector<E> b)
void	synchronize()	Synchronizes GPU execution (waits for all kernels to complete).
Matrix<E>	toDense(Matrix<E> a)	Converts a sparse matrix to its dense representation.
E	trace(Matrix<E> a)

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface AlgorithmProvider

description, getMetadata

Methods inherited from interface Backend

getAlgorithmProviders, getStatusMessage

Methods inherited from interface ComputeBackend

createBackend, getDescription, supportsComplexNumbers, supportsFloatingPoint, supportsParallelOps

Methods inherited from interface LinearAlgebraProvider

acos, acosh, add, angle, asin, asinh, atan, atanh, cbrt, cholesky, configure, cos, cosh, cross, determinant, eigen, exp, getEnvironmentInfo, inverse, log, log10, lu, multiply, multiply, normalize, pow, projection, qr, scale, score, sin, sinh, solve, solve, solve, solveTriangular, sqrt, subtract, svd, tan, tanh, transpose

Methods inherited from interface SparseLinearAlgebraProvider

conditionNumber, getAlgorithmType, gmres, rank

Constructor Details

NativeCUDASparseLinearAlgebraDoubleBackend

public NativeCUDASparseLinearAlgebraDoubleBackend()

Method Details

isLoaded

public boolean isLoaded()

Description copied from interface: NativeBackend

Checks if the native library is loaded and available.

Specified by:

isLoaded in interface NativeBackend

Returns:

true if the native library has been successfully loaded

getNativeLibraryName

public String getNativeLibraryName()

Description copied from interface: NativeBackend

Returns the name of the native library this backend depends on.

Specified by:

getNativeLibraryName in interface NativeBackend

Returns:

library name (e.g., "openblas", "cuda", "fftw3")

isAvailable

public boolean isAvailable()

Description copied from interface: AlgorithmProvider

Checks if the provider is available in the current environment.

Specified by:

isAvailable in interface AlgorithmProvider

Specified by:

isAvailable in interface Backend

isExplicitlyDisabled

public boolean isExplicitlyDisabled()

Description copied from interface: Backend

Checks if this backend is specifically disabled via a system property: "episteme.backend.disable." + getId().

It also checks specialized properties like "episteme.linearalgebra.disable." + getId().

Specified by:

isExplicitlyDisabled in interface Backend

Returns:

true if explicitly disabled

isCompatible

public boolean isCompatible(Ring<?> ring)

Description copied from interface: LinearAlgebraProvider

Checks if this provider is compatible with the given ring.

Specified by:

isCompatible in interface LinearAlgebraProvider<E extends FieldElement<E>>

getId

public String getId()

Description copied from interface: Backend

Returns the unique identifier for this backend (e.g., "javafx", "jmol", "cuda").

Specified by:

getId in interface Backend

Specified by:

getId in interface ComputeBackend

getName

public String getName()

Description copied from interface: Backend

Returns the display name for UI presentation.

Specified by:

getName in interface AlgorithmProvider

Specified by:

getName in interface Backend

Specified by:

getName in interface LinearAlgebraProvider<E extends FieldElement<E>>

getPriority

public int getPriority()

Description copied from interface: LinearAlgebraProvider

Priority of this provider (higher means more preferred). Used for automatic backend selection.

Specified by:

getPriority in interface AlgorithmProvider

Specified by:

getPriority in interface Backend

Specified by:

getPriority in interface LinearAlgebraProvider<E extends FieldElement<E>>

shutdown

public void shutdown()

Description copied from interface: AlgorithmProvider

Called when the provider is no longer needed (e.g., application shutdown). Use this to release local resources or close native segments.

Specified by:

shutdown in interface AlgorithmProvider

Specified by:

shutdown in interface Backend

multiply

public Vector<E> multiply(Matrix<E> a,
 Vector<E> x)

Specified by:

multiply in interface LinearAlgebraProvider<E extends FieldElement<E>>

toDense

public Matrix<E> toDense(Matrix<E> a)

Description copied from interface: SparseLinearAlgebraProvider

Converts a sparse matrix to its dense representation.

Specified by:

toDense in interface SparseLinearAlgebraProvider<E extends FieldElement<E>>

fromDense

public Matrix<E> fromDense(Matrix<E> a)

Description copied from interface: SparseLinearAlgebraProvider

Converts a dense matrix to its sparse representation.

Specified by:

fromDense in interface SparseLinearAlgebraProvider<E extends FieldElement<E>>

solve

public Vector<E> solve(Matrix<E> a,
 Vector<E> b)

Specified by:

solve in interface LinearAlgebraProvider<E extends FieldElement<E>>

conjugateGradient

public Vector<E> conjugateGradient(Matrix<E> a,
 Vector<E> b,
 Vector<E> x0,
 E tolerance,
 int maxIterations)

Description copied from interface: SparseLinearAlgebraProvider

Solves Ax = b using Conjugate Gradient method.

Specified by:

conjugateGradient in interface SparseLinearAlgebraProvider<E extends FieldElement<E>>

bicgstab

public Vector<E> bicgstab(Matrix<E> a,
 Vector<E> b,
 Vector<E> x0,
 E tolerance,
 int maxIterations)

Description copied from interface: SparseLinearAlgebraProvider

Solves Ax = b using BiCGSTAB (BiConjugate Gradient Stabilized) method.

Specified by:

bicgstab in interface SparseLinearAlgebraProvider<E extends FieldElement<E>>

add

public Vector<E> add(Vector<E> a,
 Vector<E> b)

Specified by:

add in interface LinearAlgebraProvider<E extends FieldElement<E>>

subtract

public Vector<E> subtract(Vector<E> a,
 Vector<E> b)

Specified by:

subtract in interface LinearAlgebraProvider<E extends FieldElement<E>>

dot

public E dot(Vector<E> a,
 Vector<E> b)

Specified by:

dot in interface LinearAlgebraProvider<E extends FieldElement<E>>

norm

public E norm(Vector<E> a)

Specified by:

norm in interface LinearAlgebraProvider<E extends FieldElement<E>>

trace

public E trace(Matrix<E> a)

Specified by:

trace in interface LinearAlgebraProvider<E extends FieldElement<E>>

conjugateTranspose

public Matrix<E> conjugateTranspose(Matrix<E> a)

Specified by:

conjugateTranspose in interface LinearAlgebraProvider<E extends FieldElement<E>>

close

public void close()

Specified by:

close in interface AutoCloseable

Specified by:

close in interface LinearAlgebraProvider<E extends FieldElement<E>>

getAcceleratorType

public 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

getType

public 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

createContext

public ExecutionContext createContext()

Description copied from interface: ComputeBackend

Creates an execution context for running operations.

The context should be used within a try-with-resources block to ensure proper cleanup of resources.

Specified by:

createContext in interface ComputeBackend

Returns:

new execution context

getDevices

public GPUBackend.DeviceInfo[] getDevices()

Description copied from interface: GPUBackend

Returns information about available GPU devices.

Specified by:

getDevices in interface GPUBackend

selectDevice

public void selectDevice(int deviceId)

Description copied from interface: GPUBackend

Selects a GPU device for computation.

Specified by:

selectDevice in interface GPUBackend

Parameters:

deviceId - Device index (0-based)

allocateGPUMemory

public long allocateGPUMemory(long sizeBytes)

Description copied from interface: GPUBackend

Allocates GPU memory and returns a handle.

Specified by:

allocateGPUMemory in interface GPUBackend

Parameters:

sizeBytes - Size in bytes

Returns:

GPU memory handle (opaque pointer)

copyToGPU

public void copyToGPU(long gpuHandle,
 DoubleBuffer hostBuffer,
 long sizeBytes)

Description copied from interface: GPUBackend

Copies data from host to GPU.

Specified by:

copyToGPU in interface GPUBackend

Parameters:

gpuHandle - GPU memory handle

hostBuffer - Host buffer

sizeBytes - Number of bytes to copy

copyFromGPU

public void copyFromGPU(long gpuHandle,
 DoubleBuffer hostBuffer,
 long sizeBytes)

Description copied from interface: GPUBackend

Copies data from GPU to host.

Specified by:

copyFromGPU in interface GPUBackend

Parameters:

gpuHandle - GPU memory handle

hostBuffer - Host buffer

sizeBytes - Number of bytes to copy

freeGPUMemory

public void freeGPUMemory(long gpuHandle)

Description copied from interface: GPUBackend

Frees GPU memory.

Specified by:

freeGPUMemory in interface GPUBackend

Parameters:

gpuHandle - GPU memory handle

synchronize

public void synchronize()

Description copied from interface: GPUBackend

Synchronizes GPU execution (waits for all kernels to complete).

Specified by:

synchronize in interface GPUBackend

matrixMultiply

public void matrixMultiply(DoubleBuffer A,
 DoubleBuffer B,
 DoubleBuffer C,
 int m,
 int n,
 int k)

Description copied from interface: GPUBackend

Performs matrix multiplication C = A * B on the GPU. Dimensions: A(m x k), B(k x n), C(m x n).

Specified by:

matrixMultiply in interface GPUBackend