Package org.apache.druid.segment.indexing

Interface TuningConfig

    • Field Detail

      • DEFAULT_LOG_PARSE_EXCEPTIONS

        static final boolean DEFAULT_LOG_PARSE_EXCEPTIONS
        See Also:
        Constant Field Values

      • DEFAULT_APPENDABLE_INDEX

        static final org.apache.druid.segment.incremental.AppendableIndexSpec DEFAULT_APPENDABLE_INDEX

      • DEFAULT_MAX_PARSE_EXCEPTIONS

        static final int DEFAULT_MAX_PARSE_EXCEPTIONS
        See Also:
        Constant Field Values

      • DEFAULT_MAX_SAVED_PARSE_EXCEPTIONS

        static final int DEFAULT_MAX_SAVED_PARSE_EXCEPTIONS
        See Also:
        Constant Field Values

      • DEFAULT_MAX_ROWS_IN_MEMORY_BATCH

        static final int DEFAULT_MAX_ROWS_IN_MEMORY_BATCH
        For batch ingestion, we want to maximize throughput by minimizing the number of incremental persists. The limit here is really a safety: in case we have a large number of very small rows, we don't want to get overwhelmed by per-row overheads. Mostly, we rely on the bytes limit getMaxBytesInMemory().
        See Also:
        Constant Field Values

      • DEFAULT_MAX_ROWS_IN_MEMORY_REALTIME

        static final int DEFAULT_MAX_ROWS_IN_MEMORY_REALTIME
        For realtime ingestion, we want to balance ingestion throughput and query performance. Since queries on in-memory data are slower due to using IncrementalIndex instead of QueryableIndex, we cap the row count of in-memory data.
        See Also:
        Constant Field Values

      • DEFAULT_SKIP_BYTES_IN_MEMORY_OVERHEAD_CHECK

        static final boolean DEFAULT_SKIP_BYTES_IN_MEMORY_OVERHEAD_CHECK
        See Also:
        Constant Field Values

      • DEFAULT_AWAIT_SEGMENT_AVAILABILITY_TIMEOUT_MILLIS

        static final long DEFAULT_AWAIT_SEGMENT_AVAILABILITY_TIMEOUT_MILLIS
        See Also:
        Constant Field Values

    • Method Detail

      • getAppendableIndexSpec

        org.apache.druid.segment.incremental.AppendableIndexSpec getAppendableIndexSpec()
        The incremental index implementation to use

      • getMaxRowsInMemory

        int getMaxRowsInMemory()
        Maximum number of rows in memory before persisting to local storage

      • getMaxBytesInMemory

        long getMaxBytesInMemory()
        Maximum number of bytes (estimated) to store in memory before persisting to local storage

      • getMaxBytesInMemoryOrDefault

        default long getMaxBytesInMemoryOrDefault()
        Maximum number of bytes (estimated) to store in memory before persisting to local storage. If getMaxBytesInMemory() returns 0, the appendable index default will be used.

      • getPartitionsSpec

        org.apache.druid.indexer.partitions.PartitionsSpec getPartitionsSpec()

      • getIndexSpec

        org.apache.druid.segment.IndexSpec getIndexSpec()

      • getIndexSpecForIntermediatePersists

        org.apache.druid.segment.IndexSpec getIndexSpecForIntermediatePersists()