ai.rapids.cudf.nvcomp

Class LZ4Compressor

java.lang.Object

ai.rapids.cudf.nvcomp.LZ4Compressor

public class LZ4Compressor
extends Object

Single-buffer compressor implementing LZ4

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	LZ4Compressor.Configuration LZ4 compression settings corresponding to a chunk size

Constructor Summary

Constructors

Constructor and Description
LZ4Compressor()

Method Summary

Modifier and Type	Method and Description
static long	compress(BaseDeviceMemoryBuffer input, CompressionType inputType, long chunkSize, BaseDeviceMemoryBuffer tempBuffer, BaseDeviceMemoryBuffer output, Cuda.Stream stream) Synchronously compress a buffer with LZ4.
static void	compressAsync(DeviceMemoryBuffer compressedSizeOutputBuffer, BaseDeviceMemoryBuffer input, CompressionType inputType, long chunkSize, BaseDeviceMemoryBuffer tempBuffer, BaseDeviceMemoryBuffer output, Cuda.Stream stream) Asynchronously compress a buffer with LZ4.
static LZ4Compressor.Configuration	configure(long chunkSize, long uncompressedSize) Get the compression configuration necessary for a particular chunk size.

Methods inherited from class java.lang.Object

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

Constructor Detail

LZ4Compressor

public LZ4Compressor()

Method Detail

configure

public static LZ4Compressor.Configuration configure(long chunkSize,
                                                    long uncompressedSize)

Get the compression configuration necessary for a particular chunk size.

Parameters:

chunkSize - size of an LZ4 chunk in bytes

uncompressedSize - total size of the uncompressed data

Returns:

compression configuration for the specified chunk size

compress

public static long compress(BaseDeviceMemoryBuffer input,
                            CompressionType inputType,
                            long chunkSize,
                            BaseDeviceMemoryBuffer tempBuffer,
                            BaseDeviceMemoryBuffer output,
                            Cuda.Stream stream)

Synchronously compress a buffer with LZ4.

Parameters:

input - buffer to compress

inputType - type of data within the buffer

chunkSize - compression chunk size to use

tempBuffer - temporary storage space

output - buffer that will contain the compressed result

stream - CUDA stream to use

Returns:

size of the resulting compressed data stored to the output buffer

compressAsync

public static void compressAsync(DeviceMemoryBuffer compressedSizeOutputBuffer,
                                 BaseDeviceMemoryBuffer input,
                                 CompressionType inputType,
                                 long chunkSize,
                                 BaseDeviceMemoryBuffer tempBuffer,
                                 BaseDeviceMemoryBuffer output,
                                 Cuda.Stream stream)

Asynchronously compress a buffer with LZ4. The compressed size output buffer must be pinned memory for this operation to be truly asynchronous. Note that the caller must synchronize on the specified CUDA stream in order to safely examine the compressed output size!

Parameters:

compressedSizeOutputBuffer - device memory where the compressed output size will be stored

input - buffer to compress

inputType - type of data within the buffer

chunkSize - compression chunk size to use

tempBuffer - temporary storage space

output - buffer that will contain the compressed result

stream - CUDA stream to use