...
The Apache Ignite can support cache rebalancing as transferring partition files using zero copy algorithm [1] based on an extension of communication SPI and Java NIO API. When the partition file has been transferred to the demander node there are two a few possible approaches can be implemented to preload entries from particular partition file.
The demander node will initialize a preloaded partition file as a new PageStore, make this storage up-to-date by applying previously saved async cache operations and then hot swap it under the checkpoint write lock.
Disadvantages:
Demander node first under checkpoint write lock must swap cache data storage with the temporary one to perform recovery operations under original cache data storage. After partition file has been received from the Supplier node there are to possible cases to make this partition file up-to-date.
Disadvantages:
The demander node will use a preloaded patition file as a new source of cache data entries to load.
Disadvantages:
In the process of balancing data:
The whole process is described in terms of rebalance single cache group and partition files would be rebalanced one-by-one:
In terms of a high-level abstraction, Apache Ignite must support the features described below.
Partition preloader must support cache partition file relocation from one cluster node to another (the zero copy algorithm [1] assume to be used by default). To achieve this, the file transfer machinery must be implemented at Apache Ignite over Communication SPI.
Ignite's Comminication SPI must support to:
After partition is received the historical rebalance must be initiated to load other cache updates.
The swapped temporary storage will log all the cache updates to the temporary WAL storage (per each partition) for further applying them to the corresponding partition file. While the Demander is being receive partition files it must save sequentially all cache entries corresponding to the MOVING partition into a new temporary storage. These entries will be applied later one by one on the newly received cache partition file. All asynchronous operations will be enrolled to the end of temporary storage during storage reads until it becomes fully read. The file-based FIFO approach assumes to be used by this process.
The temporary storage is chosen to be WAL-based. The storage must support to:
Expected problems to be solved
The demander node will use a preloaded patition file as a new source of cache data entries to load.
Disadvantages:
In the process of balancing data:
The whole process is described in terms of rebalancing a single partition file of a cache group. All the other partitions would be rebalanced one-by-one.
In terms of a high-level abstraction, Apache Ignite must support the features described below.
The node partition preloader machinery download cache partition files from cluster nodes which owns desired partitions (the zero copy algorithm [1] assume to be used by default). To achieve this, the file transmission process must be implemented at Apache Ignite over Communication SPI.
IThe Comminication SPI must support to:
Code Block | ||||||||
---|---|---|---|---|---|---|---|---|
| ||||||||
public interface CommunicationListenerEx<T extends Serializable> extends EventListener { /** * @param nodeId Remote node id. * @param initMsg Init channel message. * @param channel Locally created channel endpoint. */ public void onChannelOpened(UUID nodeId, Message initMsg, Channel channel); } |
...
The file transmission IO manager must support to:
Code Block | ||||||||
---|---|---|---|---|---|---|---|---|
| ||||||||
/** * */ public interface FileWriter extends CloseableTransmissionHandler { /** * @param fileerr The sourceerr fileof tofail sendhandling atprocess. */ @param offset The positionpublic to start at.void onException(UUID nodeId, Throwable err); /** * @param count The number of bytes to transfernodeId Remote node id from which request has been received. * @param paramsfileMeta TheFile additional transfer file description keysmeta info. * @param@return plcAbsolute Thepathname policydenoting of handling data on remotea file. */ @throws IgniteCheckedException If fails. public String filePath(UUID nodeId, */TransmissionMeta fileMeta); public void write(/** * <em>Chunk handler</em> Filerepresents file, by itself the way of input data stream long offset, processing. * It accepts within longeach count, chunk a {@link ByteBuffer} with data from input Map<String,for Serializable>further params,processing. * ReadPolicy * plc @param nodeId Remote node )id throws IgniteCheckedException; } | ||||||||
Code Block | ||||||||
| ||||||||
/** * */ public interface TransmitSessionHandler { /**from which request has been received. * @param initMeta Initial handler meta info. * @param@return nodeIdInstance Theof remotechunk nodehandler idto connectedprocess from. incoming data * @param sessionId The unique session idby chunks. */ public voidConsumer<ByteBuffer> beginchunkHandler(UUID nodeId, StringTransmissionMeta sessionIdinitMeta); /** * @return<em>File Thehandler</em> instancerepresents ofby readitself handlerthe toway processof incominginput data bystream chunksprocessing. All the data */will public ChunkHandler chunkHandler(); /** * @return The intance of read handler to process incoming data like the {@link FileChannel} manner* be processed under the hood using zero-copy transferring algorithm and only start file processing and * the end of processing will be provided. */ public FileHandler fileHandler(); /** * @param nodeId Remote node id from which request has been received. * The@param endinitMeta ofInitial sessionhandler transmissionmeta processinfo. */ @return Intance of publicread void end(); /** * @param cause The cause of fail handling processhandler to process incoming data like the {@link FileChannel} manner. */ public voidConsumer<File> onException(Throwable causefileHandler(UUID nodeId, TransmissionMeta initMeta); } |
Code Block | ||||||||
---|---|---|---|---|---|---|---|---|
| ||||||||
/** * */ public interface FileHandlerpublic class TransmissionSender implements Closeable { /** * @param namefile TheSource file to namesend transferto fromremote. * @param positionparams TheAdditional start position pointer of download transfer file indescription original sourcekeys. * @param countplc TotalThe countpolicy of byteshandling readeddata from the original sourceon remote. * @param@throws paramsIgniteCheckedException The additional transfer file description paramsIf fails. */ @return The absolute pathnamepublic stringvoid denotingsend( the file or {@code null} if there is noFile sense.file, * @throws IOException If fails.Map<String, Serializable> params, */ TransmissionPolicy public String begin(String name, long position, long count, Map<String, Serializable> params) throws IOException; plc ) throws IgniteCheckedException, InterruptedException, IOException { /** * @param file The file with fully downloaded data into.send(file, 0, file.length(), params, plc); } /** * @param paramsfile TheSource additionalfile transferto filesend descriptionto paramsremote. */ @param plc The publicpolicy voidof end(File file, Map<String, Serializable> params); } | ||||||||
Code Block | ||||||||
| ||||||||
/** * */ public interface ChunkHandler { handling data on remote. * @throws IgniteCheckedException If fails. */** public *void @paramsend( name The file name on remote. File file, * @param position The start position pointerTransmissionPolicy ofplc downloading file in original) source. throws IgniteCheckedException, InterruptedException, IOException { * @param count Total count of bytes to read from the original source.send(file, 0, file.length(), new HashMap<>(), plc); *} @param params The additional transfer file description params. /** * @return@param Thefile sizeSource offile of {@link ByteBuffer}to send to read the input channel intoremote. * @throws IOException If fails@param offset Position to start trasfer at. */ @param cnt Number publicof intbytes begin(String name, long position, long count, Map<String, Serializable> params) throws IOException; /**to transfer. * @param params Additional transfer file description keys. * @param buffplc The datapolicy filled buffer. * @return {@code true} if the chunk of data have been successfully acceptedof handling data on remote. * @throws IOExceptionIgniteCheckedException If fails. */ public booleanvoid chunksend(ByteBuffer buff) throws IOException; File /**file, * @param params Thelong additionaloffset, handling channel description params. long */cnt, public void end(Map<String, Serializable> params); } |
When the supplier node receives the cache partition file demand request it will send the file over the CommunicationSpi. The cache partition file can be concurrently updated by checkpoint thread during its transmission. To guarantee the file consistency Сheckpointer must use Copy-on-Write [3] tehnique and save a copy of updated chunk into the temporary file.
While the demander node is in the partition file transmission state it must save sequentially all cache entries corresponding to the MOVING partition into a new temporary storage. These entries will be applied later one by one on the newly received cache partition file. All asynchronous operations will be enrolled to the end of temporary storage during storage reads until it becomes fully read. The file-based FIFO approach assumes to be used by this process.
The temporary storage is chosen to be WAL-based. The storage must support to:
Expected problems to be solved
...
,
TransmissionPolicy plc
) throws IgniteCheckedException, InterruptedException, IOException {
// Impl.
}
}
|
When the supplier node receives the cache partition file demand request it will send the file over the CommunicationSpi. The cache partition file can be concurrently updated by checkpoint thread during its transmission. To guarantee the file consistency Сheckpointer must use Copy-on-Write [3] tehnique and save a copy of updated chunk into the temporary file.
...
The node is ready to become partition owner when partition data is rebalanced and cache indexes are ready. For the message-based cluster rebalancing approach indexes are rebuilding simultaneously with cache data loading. For the file-based rebalancing approach, the index rebuild procedure must be finished before the partition state is set to the OWNING state.
procedure must be finished before the partition state is set to the OWNING state.
Ignite doesn't provide any recovery guarantees for the partitions with the MOVING state. The cache partitions will be fully loaded when the next rebalance procedure occurs.
The node which is beeing rebalancing left the cluster. For such nodes WAL is always disabled (all partitions have MOVING state due to this node is new for the cluster and has no cache data).
Since WAL is disabled we can guarantee that all operations with loaded partition files are safe to be done (renaming partition files, applying async updates) due to a cache directory will be fully dropped on recoveryApache Ignite doesn't provide any recovery guarantees for the partitions with the MOVING state. The cache partitions will be fully loaded when the next rebalance procedure occurs.
Each topology change event JOIN/LEFT/FAILED may or may not change cache affinity assignments of currently rebalacning caches. If assignments is not changed and the node is still needs partitions being rebalanced we can continue the current rebalance process (see for details IGNITE-7165).
...
To provide basic recovery guarantees we must to:
...
Recovery from different stages:
The SSL must be disabled to take an advantage of Java NIO zero-copy file transmission using FileChannel#transferTo method. If we need to use SSL the file must be splitted on chunks the same way to send them over the socket channel with ByteBuffer. As the SSL engine generally needs a direct ByteBuffer to do encryption we can't avoid copying buffer payload from the kernel level to the application level.
...