org.rocksdb

Class BlockBasedTableConfig

java.lang.Object

org.rocksdb.TableFormatConfig

org.rocksdb.BlockBasedTableConfig

public class BlockBasedTableConfig
extends TableFormatConfig

The config for plain table sst format. BlockBasedTable is a RocksDB's default SST file format.

Constructor Summary

Constructors

Constructor and Description
BlockBasedTableConfig()

Method Summary

Modifier and Type	Method and Description
int	blockCacheCompressedNumShardBits() Controls the number of shards for the block compressed cache.
long	blockCacheCompressedSize() Size of compressed block cache.
long	blockCacheSize()
int	blockRestartInterval()
long	blockSize()
int	blockSizeDeviation()
boolean	cacheIndexAndFilterBlocks() Indicating if we'd put index/filter blocks to the block cache.
int	cacheNumShardBits() Returns the number of shard bits used in the block cache.
ChecksumType	checksumType()
int	formatVersion()
boolean	hashIndexAllowCollision() Influence the behavior when kHashSearch is used.
IndexType	indexType()
protected long	newTableFactoryHandle() This function should only be called by Options.setTableFormatConfig(), which will create a c++ shared-pointer to the c++ TableFactory that associated with the Java TableFormatConfig.
boolean	noBlockCache()
boolean	pinL0FilterAndIndexBlocksInCache() Indicating if we'd like to pin L0 index/filter blocks to the block cache.
BlockBasedTableConfig	setBlockCacheCompressedNumShardBits(int blockCacheCompressedNumShardBits) Controls the number of shards for the block compressed cache.
BlockBasedTableConfig	setBlockCacheCompressedSize(long blockCacheCompressedSize) Size of compressed block cache.
BlockBasedTableConfig	setBlockCacheSize(long blockCacheSize) Set the amount of cache in bytes that will be used by RocksDB.
BlockBasedTableConfig	setBlockRestartInterval(int restartInterval) Set block restart interval
BlockBasedTableConfig	setBlockSize(long blockSize) Approximate size of user data packed per block.
BlockBasedTableConfig	setBlockSizeDeviation(int blockSizeDeviation) This is used to close a block before it reaches the configured 'block_size'.
BlockBasedTableConfig	setCacheIndexAndFilterBlocks(boolean cacheIndexAndFilterBlocks) Indicating if we'd put index/filter blocks to the block cache.
BlockBasedTableConfig	setCacheNumShardBits(int blockCacheNumShardBits) Controls the number of shards for the block cache.
BlockBasedTableConfig	setChecksumType(ChecksumType checksumType) Sets the checksum type to be used with this table.
BlockBasedTableConfig	setFilter(Filter filter) Use the specified filter policy to reduce disk reads.
BlockBasedTableConfig	setFormatVersion(int formatVersion) We currently have three versions:
BlockBasedTableConfig	setHashIndexAllowCollision(boolean hashIndexAllowCollision) Influence the behavior when kHashSearch is used.
BlockBasedTableConfig	setIndexType(IndexType indexType) Sets the index type to used with this table.
BlockBasedTableConfig	setNoBlockCache(boolean noBlockCache) Disable block cache.
BlockBasedTableConfig	setPinL0FilterAndIndexBlocksInCache(boolean pinL0FilterAndIndexBlocksInCache) Indicating if we'd like to pin L0 index/filter blocks to the block cache.
BlockBasedTableConfig	setWholeKeyFiltering(boolean wholeKeyFiltering) If true, place whole keys in the filter (not just prefixes).
boolean	wholeKeyFiltering()

Methods inherited from class java.lang.Object

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

Constructor Detail

BlockBasedTableConfig

public BlockBasedTableConfig()

Method Detail

setNoBlockCache

public BlockBasedTableConfig setNoBlockCache(boolean noBlockCache)

Disable block cache. If this is set to true, then no block cache should be used, and the block_cache should point to a nullptr object. Default: false

Parameters:

noBlockCache - if use block cache

Returns:

the reference to the current config.

noBlockCache

public boolean noBlockCache()

Returns:

if block cache is disabled

setBlockCacheSize

public BlockBasedTableConfig setBlockCacheSize(long blockCacheSize)

Set the amount of cache in bytes that will be used by RocksDB. If cacheSize is non-positive, then cache will not be used. DEFAULT: 8M

Parameters:

blockCacheSize - block cache size in bytes

Returns:

the reference to the current config.

blockCacheSize

public long blockCacheSize()

Returns:

block cache size in bytes

setCacheNumShardBits

public BlockBasedTableConfig setCacheNumShardBits(int blockCacheNumShardBits)

Controls the number of shards for the block cache. This is applied only if cacheSize is set to non-negative.

Parameters:

blockCacheNumShardBits - the number of shard bits. The resulting number of shards would be 2 ^ numShardBits. Any negative number means use default settings."

Returns:

the reference to the current option.

cacheNumShardBits

public int cacheNumShardBits()

Returns the number of shard bits used in the block cache. The resulting number of shards would be 2 ^ (returned value). Any negative number means use default settings.

Returns:

the number of shard bits used in the block cache.

setBlockSize

public BlockBasedTableConfig setBlockSize(long blockSize)

Approximate size of user data packed per block. Note that the block size specified here corresponds to uncompressed data. The actual size of the unit read from disk may be smaller if compression is enabled. This parameter can be changed dynamically. Default: 4K

Parameters:

blockSize - block size in bytes

Returns:

the reference to the current config.

blockSize

public long blockSize()

Returns:

block size in bytes

setBlockSizeDeviation

public BlockBasedTableConfig setBlockSizeDeviation(int blockSizeDeviation)

This is used to close a block before it reaches the configured 'block_size'. If the percentage of free space in the current block is less than this specified number and adding a new record to the block will exceed the configured block size, then this block will be closed and the new record will be written to the next block. Default is 10.

Parameters:

blockSizeDeviation - the deviation to block size allowed

Returns:

the reference to the current config.

blockSizeDeviation

public int blockSizeDeviation()

Returns:

the hash table ratio.

setBlockRestartInterval

public BlockBasedTableConfig setBlockRestartInterval(int restartInterval)

Set block restart interval

Parameters:

restartInterval - block restart interval.

Returns:

the reference to the current config.

blockRestartInterval

public int blockRestartInterval()

Returns:

block restart interval

setWholeKeyFiltering

public BlockBasedTableConfig setWholeKeyFiltering(boolean wholeKeyFiltering)

If true, place whole keys in the filter (not just prefixes). This must generally be true for gets to be efficient. Default: true

Parameters:

wholeKeyFiltering - if enable whole key filtering

Returns:

the reference to the current config.

wholeKeyFiltering

public boolean wholeKeyFiltering()

Returns:

if whole key filtering is enabled

setFilter

public BlockBasedTableConfig setFilter(Filter filter)

Use the specified filter policy to reduce disk reads. Filter should not be disposed before options instances using this filter is disposed. If AbstractNativeReference.dispose() function is not called, then filter object will be GC'd automatically. Filter instance can be re-used in multiple options instances.

Parameters:

filter - Filter Filter Policy java instance.

Returns:

the reference to the current config.

cacheIndexAndFilterBlocks

public boolean cacheIndexAndFilterBlocks()

Indicating if we'd put index/filter blocks to the block cache. If not specified, each "table reader" object will pre-load index/filter block during table initialization.

Returns:

if index and filter blocks should be put in block cache.

setCacheIndexAndFilterBlocks

public BlockBasedTableConfig setCacheIndexAndFilterBlocks(boolean cacheIndexAndFilterBlocks)

Indicating if we'd put index/filter blocks to the block cache. If not specified, each "table reader" object will pre-load index/filter block during table initialization.

Parameters:

cacheIndexAndFilterBlocks - and filter blocks should be put in block cache.

Returns:

the reference to the current config.

pinL0FilterAndIndexBlocksInCache

public boolean pinL0FilterAndIndexBlocksInCache()

Indicating if we'd like to pin L0 index/filter blocks to the block cache. If not specified, defaults to false.

Returns:

if L0 index and filter blocks should be pinned to the block cache.

setPinL0FilterAndIndexBlocksInCache

public BlockBasedTableConfig setPinL0FilterAndIndexBlocksInCache(boolean pinL0FilterAndIndexBlocksInCache)

Indicating if we'd like to pin L0 index/filter blocks to the block cache. If not specified, defaults to false.

Parameters:

pinL0FilterAndIndexBlocksInCache - pin blocks in block cache

Returns:

the reference to the current config.

hashIndexAllowCollision

public boolean hashIndexAllowCollision()

Influence the behavior when kHashSearch is used. if false, stores a precise prefix to block range mapping if true, does not store prefix and allows prefix hash collision (less memory consumption)

Returns:

if hash collisions should be allowed.

setHashIndexAllowCollision

public BlockBasedTableConfig setHashIndexAllowCollision(boolean hashIndexAllowCollision)

Influence the behavior when kHashSearch is used. if false, stores a precise prefix to block range mapping if true, does not store prefix and allows prefix hash collision (less memory consumption)

Parameters:

hashIndexAllowCollision - points out if hash collisions should be allowed.

Returns:

the reference to the current config.

blockCacheCompressedSize

public long blockCacheCompressedSize()

Size of compressed block cache. If 0, then block_cache_compressed is set to null.

Returns:

size of compressed block cache.

setBlockCacheCompressedSize

public BlockBasedTableConfig setBlockCacheCompressedSize(long blockCacheCompressedSize)

Size of compressed block cache. If 0, then block_cache_compressed is set to null.

Parameters:

blockCacheCompressedSize - of compressed block cache.

Returns:

the reference to the current config.

blockCacheCompressedNumShardBits

public int blockCacheCompressedNumShardBits()

Controls the number of shards for the block compressed cache. This is applied only if blockCompressedCacheSize is set to non-negative.

Returns:

numShardBits the number of shard bits. The resulting number of shards would be 2 ^ numShardBits. Any negative number means use default settings.

setBlockCacheCompressedNumShardBits

public BlockBasedTableConfig setBlockCacheCompressedNumShardBits(int blockCacheCompressedNumShardBits)

Controls the number of shards for the block compressed cache. This is applied only if blockCompressedCacheSize is set to non-negative.

Parameters:

blockCacheCompressedNumShardBits - the number of shard bits. The resulting number of shards would be 2 ^ numShardBits. Any negative number means use default settings."

Returns:

the reference to the current option.

setChecksumType

public BlockBasedTableConfig setChecksumType(ChecksumType checksumType)

Sets the checksum type to be used with this table.

Parameters:

checksumType - ChecksumType value.

Returns:

the reference to the current option.

checksumType

public ChecksumType checksumType()

Returns:

the currently set checksum type

setIndexType

public BlockBasedTableConfig setIndexType(IndexType indexType)

Sets the index type to used with this table.

Parameters:

indexType - IndexType value

Returns:

the reference to the current option.

indexType

public IndexType indexType()

Returns:

the currently set index type

setFormatVersion

public BlockBasedTableConfig setFormatVersion(int formatVersion)

We currently have three versions:

• 0 - This version is currently written out by all RocksDB's versions by default. Can be read by really old RocksDB's. Doesn't support changing checksum (default is CRC32).
• 1 - Can be read by RocksDB's versions since 3.0. Supports non-default checksum, like xxHash. It is written by RocksDB when BlockBasedTableOptions::checksum is something other than kCRC32c. (version 0 is silently upconverted)
• 2 - Can be read by RocksDB's versions since 3.10. Changes the way we encode compressed blocks with LZ4, BZip2 and Zlib compression. If you don't plan to run RocksDB before version 3.10, you should probably use this.

This option only affects newly written tables. When reading existing tables, the information about version is read from the footer.

Parameters:

formatVersion - integer representing the version to be used.

Returns:

the reference to the current option.

formatVersion

public int formatVersion()

Returns:

the currently configured format version. See also: setFormatVersion(int).

newTableFactoryHandle

protected long newTableFactoryHandle()

Description copied from class: TableFormatConfig

This function should only be called by Options.setTableFormatConfig(), which will create a c++ shared-pointer to the c++ TableFactory that associated with the Java TableFormatConfig.

Specified by:

newTableFactoryHandle in class TableFormatConfig

Returns:

native handle address to native table instance.