In the case of a HiRDB/Single Server, specify the name of the local RD node in the pd_node_name operand of the single server definition. In the case of a HiRDB/Parallel Server, specify the name of the local RD node in the pd_node_name operand of the front-end server definition.
When the distributed client facility is to be used, specify whether or not batch retrieval is to be performed. Specify in the PDRDABLKF operand the number of rows that are to be subject to batch retrieval. Batch retrieval means transferring from the distributed server to the HiRDB system multiple rows of retrieval results in the batch mode. Specifying the PDRDABLKF operand reduces the number of data transfer operations, thereby reducing communications overhead and retrieval time. For details on the PDRDABLKF operand, see the manual HiRDB Version 8 UAP Development Guide.
When batch retrieval is used, more data is received than is customary during retrieval processing. Therefore, if batch retrieval is not possible due to limitations with respect to the DF/UX transfer buffer, a FETCH statement may result in an error. For details on the error handling procedures and the maximum amount of receive data, see the manual Distributed Database System DF/UX.
If the distributed server DBMS does not support batch retrieval, regular retrieval (single-row retrieval) is executed.