: (Long) Starting byte of the header blocks
: (Long) Starting byte of the data blocks
: (Long) Last byte of non-zero data (could be in a middle of a data block)
: (Long) Last byte of the data blocks (data blocks are multiple of 2880 bytes)
: (Long) Last byte of the data blocks (data blocks are multiple of 2880 bytes)
: (Long) Starting byte of the data blocks
: (Long) Last byte of non-zero data (could be in a middle of a data block)
Check whether the data block is empty based on start/end indices.
Check whether the data block is empty based on start/end indices.
(Boolean) True if the data block has zero size. False otherwise.
: (Long) Starting byte of the header blocks
Register the boundaries of the HDU in the Hadoop configuration.
Register the boundaries of the HDU in the Hadoop configuration. By doing this, we broadcast the values to the executors. It is sent as a long String, and can be read properly afterwards using retrieveBlockBoundaries. Make sure you use the same separators.
Class to hold block boundaries. These values are computed at first file scan then encoded to be broadcasted to all datanodes through the Hadoop configuration block.
: (Long) Starting byte of the header blocks
: (Long) Starting byte of the data blocks
: (Long) Last byte of non-zero data (could be in a middle of a data block)
: (Long) Last byte of the data blocks (data blocks are multiple of 2880 bytes)