CPI4345 - Temporary distributed result file &3 built for query

CPI4345
Message Text:	Temporary distributed result file &3 built for query.
Cause Text:	Temporary distributed result file &3 was created to contain the intermediate results of the query for reason code &6. The reason codes and their meanings follow: Data from member &2 of &7 &8 in library &1 was directed to other nodes. Data from member &2 of &7 &8 in library &1 was broadcast to all nodes. Either the query contains grouping fields (GROUP BY) that do not match the partitioning keys of the distributed file or the query contains grouping criteria but no grouping fields were specified or the query contains a subquery. Query contains join criteria over a distributed file and the query was processed in multiple steps. A library and member name of N indicates the data comes from a query temporary distributed file. File &3 was built on nodes: &9. It was built using partitioning keys: &10. A partitioning key of N indicates no partitioning keys were used when building the temporary distributed result file.
Recovery Text:	If the reason code is: Generally, a file is directed when the join fields do not match the partitioning keys of the distributed file. When a file is directed, the query is processed in multiple steps and processed in parallel. A temporary distributed result file is required to contain the intermediate results for each step. Generally, a file is broadcast when join fields do not match the partitioning keys of either file being joined or the join operator is not an equal operator. When a file is broadcast the query is processed in multiple steps and processed in parallel. A temporary distributed result file is required to contain the intermediate results for each step. Better performance may be achieved if grouping fields are specified that match the partitioning keys. Because the query is processed in multiple steps, a temporary distributed result file is required to contain the intermediate results for each step. See preceding message CPI4342 to determine which files were joined together. For more information about processing of distributed files, refer to the Distributed Database Programming,

CPI4345

Message Text:

Temporary distributed result file &3 built for query.

Cause Text:

Temporary distributed result file &3 was created to contain the intermediate results of the query for reason code &6. The reason codes and their meanings follow:

Data from member &2 of &7 &8 in library &1 was directed to other nodes.
Data from member &2 of &7 &8 in library &1 was broadcast to all nodes.
Either the query contains grouping fields (GROUP BY) that do not match the partitioning keys of the distributed file or the query contains grouping criteria but no grouping fields were specified or the query contains a subquery.
Query contains join criteria over a distributed file and the query was processed in multiple steps.

A library and member name of *N indicates the data comes from a query temporary distributed file.

File &3 was built on nodes: &9.

It was built using partitioning keys: &10.

A partitioning key of *N indicates no partitioning keys were used when building the temporary distributed result file.

Recovery Text:

If the reason code is:

Generally, a file is directed when the join fields do not match the partitioning keys of the distributed file. When a file is directed, the query is processed in multiple steps and processed in parallel. A temporary distributed result file is required to contain the intermediate results for each step.
Generally, a file is broadcast when join fields do not match the partitioning keys of either file being joined or the join operator is not an equal operator. When a file is broadcast the query is processed in multiple steps and processed in parallel. A temporary distributed result file is required to contain the intermediate results for each step.
Better performance may be achieved if grouping fields are specified that match the partitioning keys.
Because the query is processed in multiple steps, a temporary distributed result file is required to contain the intermediate results for each step. See preceding message CPI4342 to determine which files were joined together.

For more information about processing of distributed files, refer to the Distributed Database Programming,