Package org.apache.flink.runtime.io.network.partition.hybrid.tiered.storage
-
Interface Summary Interface Description AvailabilityNotifier AvailabilityNotifieris used to notify that the data from the specific partition and subpartition in tiered storage is available.BufferAccumulator Accumulates received records into buffers.HashSubpartitionBufferAccumulatorContext This interface is used byHashSubpartitionBufferAccumulatorto operateHashBufferAccumulator.TieredStorageMemoryManager TheTieredStorageMemoryManageris to request or recycle buffers fromLocalBufferPoolfor different memory owners, for example, the tiers, the buffer accumulator, etc.TieredStorageResource The resource (e.g., local files, remote storage files, etc.) for the Tiered Storage. -
Class Summary Class Description HashBufferAccumulator The hash implementation of theBufferAccumulator.HashSubpartitionBufferAccumulator HashSubpartitionBufferAccumulatoraccumulates the records in a subpartition.SortBufferAccumulator The sort-based implementation of theBufferAccumulator.TieredStorageConsumerClient TieredStorageConsumerClientis used to read buffer from tiered store.TieredStorageConsumerSpec Describe the different data sources inTieredStorageConsumerClient.TieredStorageMasterClient Client of the Tiered Storage used by the master.TieredStorageMemoryManagerImpl The implementation forTieredStorageMemoryManager.TieredStorageMemorySpec The memory specs for a memory owner, including the owner itself, the number of guaranteed buffers of the memory owner, etc.TieredStorageProducerClient Client of the Tiered Storage used by the producer.TieredStorageProducerMetricUpdate The metric statistics for the tiered storage producer.TieredStorageResourceRegistry A registry that maintains local or remote resources that correspond to a certain set of data in the Tiered Storage.TieredStorageSortBuffer When getting buffers, TheSortBasedDataBufferneed not recycle the read target buffer..