Salesforce

Database Information - Memory (Magic xpa 2.x)

« Go Back

Information

 
Created ByKnowledge Migration User
Approval Process StatusPublished
Objective
Description

Database Information - Memory (Magic xpa 2.x)

The Database Information parameter (in the DBMS properties, Database properties, Data Source properties, Column properties, and Index properties) lets you supply database-dependent information that Magic xpa can pass to the underlying DBMS. The use of this parameter is optional.

Multiple keywords should be delimited by one or more spaces.

Some of the keywords may not require values to be specified. All of the keywords and their values must be specified in uppercase.

Keyword

Description

CHECK_KEY=Y

Checks the data table for the existence of a unique index value as the end user attempts to input a data into a record. If a duplicate value is input by the end user into the variables constructing an index defined as Unique in Magic xpa’s Data repository, Magic xpa will issue a Duplicate Index error message and prevent further end-user processing until the input is changed. For some DBMSs, this check takes a toll in performance and it is not recommended. In such a case, the developer may decide not to use this keyword, and then the results of duplicate index value entry will depend on the DBMS behavior.

v

FILE_CHUNK_SIZE=500

Tells the engine to allocate room for a specific number of rows when loading a memory table.

A possible use for this keyword is to enhance performance. When having memory tables with large number of columns and small number of rows, reducing the chunk size will reduce memory consumption and load time.

The default value is 500. The maximum value is 500.

v

INITFILE=500

Counts the total number of concurrent memory tables a single engine is allowed to open, including sort/temporary tables, if the Database for Sort/Temporary is a memory database.

Set it to a number that reflects the total number of memory tables you will want to open.

Note that each context opens memory tables of its own so the number should be:

For a RIA server:

max number of RIA client * number of memory tables being opened for each context

(Count the memory table entries that you have in the Data repository that are really being used.)

For a non-RIA server:

number of concurrent contexts * number of memory tables being opened for each context

For an Online client:

maximum number of parallel program trees * number of memory tables that will be opened per parallel program tree

See also Why is the error message "Failed to open, table: [TEMP_PATH]\MG7832-1583.mem" appearing in the status line and mgerror.log file? troubleshooting issue.

v

Reference
Attachment 
Attachment