The following table shows the formulas for determining the file sizes required for the execution of the database reorganization utility (pdrorg):

File type	Formula (bytes)
Unload data file1 (no options specified)	n (Li + Pi) + 1200 + A + B + c 96 + D + I + F i = 1
Unload data file1 (-w option specified)	DAT or extended DAT format: c (maximum length of converted character string in column I2 + 1) n i = 1 FIX table in binary format: c (column data length i3) n i = 1 Non-FIX table in binary format: { c (column data length i3 + G) + 4 (c + 1) i = 1 } n Fixed-length character format: c (maximum length of converted character string for column i4 + crlf) n i = 1
Unload data file1 (-j option specified or during reorganization in units of schemas)5	n (Li + Pi) + i = 1 n { i = 1 m (Oij + 44) j = 1 } + 1200 + A + B + c 96 + D + I + F
LOB data unload file1	n { i = 1 m (Oij + 44) j = 1 } + 1200 + A + B + c 96 + D + I + F
Index information file	B-tree index: (K + p) n + 512 Plug-in index: (12 + X) n + 1024 These formulas are for the size of one index. If there are multiple indexes, determine the size of each index.
Process results file	1700 + number of servers storing table 500 + number of tables in schema 1000 + total number of storage RDAREAs in schema 100
Work file for sorting	Condition 1: Size of index information file + 4 n Condition 2: {Size of index information file + 4 n} 2 Condition 1 When the work buffer size specified in the sort statement E Condition 2 When the work buffer size specified in the sort statement < E E: Buffer size The buffer size obtained according to buffer-size-for-sorting for Database Reorganization Utility (pdrorg) in the manual HiRDB Version 8 Command Reference.

A: For key range partitioning: 48 + number of partitioning conditions 284

For hash partitioning: 40 + a 60

For matrix partitioning (combination of key range partitioning of the boundary value specification and hash partitioning): 48 + (number of partitioning conditions 284) + (40 + a 60)

B: n 36 (for FIX table) or (44 + c 4) n (for non-FIX table)

D: 16 + (number of LOB columns a 80)

Add the value of D only if there are LOB columns.

F: Use the following value:

d {(number of abstract-data-type attributes provided by plug-in in column i 84) + i = 1 (number of abstract-data-type LOB attributes provided by plug-in in column i a 72)} + 64 + d (84 + number of reverse generation functions i 60) i = 1

G: Number of attributes for which the return value of the reverse generation function on column I is BLOB 4

I: 136 + number of index partitions 60

Add this value when including the index.

K: Index key length

See Table 17-5 List of index key lengths. For variable-length data, keep in mind when defining the maximum length that single columns are also handled as multicolumns.

L_i: Actual length of row

Add the data length of each column to obtain the row length. If the data type of a column is BLOB, use 16 as the column length for calculating. If the data type of a column is the abstract data type provided by a plug-in, use 2 as the column length for calculating.

P_i: Data length of the abstract data type provided by a plug-in

O_ij: LOB data length

X: Value is as follows

• For abstract data type
  27 for the abstract data type stored in the LOB RDAREA
  Key length + 2 for the abstract data type of a maximum of 255 bytes of definition length
  2 for the abstract data type of 256 bytes or more of definition length
  
• 27 for the SGMLTEXT type
• 2 for FREEWORD and GEOMETRY types

a: Number of partitioned RDAREAs

c: Number of column definitions

d: Number of columns for which the abstract data type provided by a plug-in is defined

m: Number of LOB columns

n: Number of rows (for a repetition column, number of rows number of elements)

p: If all key component columns are fixed length, the value is 10; if they include a variable length, the value is 12.

crlf: Length of linefeed characters added when cr or crlf is specified in the -W option

Determine the length of linefeed characters from the following table:

-W option value		Value to be added
-W dat or -W extdat	,cr	1
	,crlf	2
	Not specified	1
-W fixtext	,cr	1
	,crlf	2
	Not specified	0

Note

When calculating the size of index information files and sort work files, if the index configuration columns are repetition columns, the number of rows to reload and n do not refer to the number of rows but to (number of rows number of elements).

¹ If the file is larger than 2 GB, take one of the following actions:

• Create multiple files, each of which is no larger than 2 GB.
• Use large files. For details about how to create large files, see 5.3 Creating HiRDB file system areas.

² Table 20-1 lists the maximum lengths of converted character strings for columns in DAT format (-W dat) or extended DAT format (-W extdat).

Table 20-1 Maximum lengths of converted character strings for columns (in DAT or extended DAT format)

Data type		Maximum length of converted character string (bytes)
Numeric data	INTEGER	11
	SMALLINT	11
	DECIMAL	31
	FLOAT	23
	SMALLFLT	23
Character string data1	CHARACTER	Defined length + 22
	VARCHAR	Actual length + 22
Mixed character string data1	MCHAR	Defined length + 22
	MVARCHAR	Actual length + 22
National character data1	NCHAR	Defined length + 22
	NVARCHAR	Actual length + 22
Date data	DATE	10
Time data	TIME	8
Date interval data	INTERVAL YEAR TO DAY	9
Time interval data	INTERVAL HOUR TO SECOND	7
Time stamp data	TIMESTAMP	19 If the number of digits for fractions of a second is not 0, add the number of digits for fractions of a second + 1.
Binary data1	BINARY	Actual length + 22

¹ If data in extended DAT format contains a double quotation mark ("), the length of the converted character string becomes longer by the number of double quotation marks.

² Two bytes are added for the enclosing brackets.

If -W dat or -W extdat is specified and sup is specified in the operand, the maximum lengths of converted character strings take effect on the columns as shown below. Note that the actual length indicates the length without the trailing consecutive spaces. For details about the space-compressed output format, see the -W option of the database reorganization utility (pdrorg) in the manual HiRDB Version 8 Command Reference.

Data type		Maximum lengths of converted character string (bytes)
Character string data	CHARACTER	Actual length + 2
Mixed character string data	MCHAR	Actual length + 2
National character data	NCHAR	Actual length + 2

³ For details about the data length, see the following tables:

• Table 17-1 List of data lengths
• Table 17-2 Data lengths for the variable-length character string type (except abstract data type and repetition columns)
• Table 17-3 Data lengths for the variable-length character string type (abstract data type)
• Table 17-4 Data lengths for the variable-length character string type (repetition columns)

⁴ Table 20-2 shows the maximum lengths of converted character strings for columns of the fixed-length character format (-W fixtext).

Table 20-2 Maximum lengths of converted character strings for columns (fixed-length character format)

Data type		Maximum lengths of converted character string (bytes)
Numeric data	INTEGER	11
	SMALLINT	6
	DECIMAL	Number of digits + 2
	FLOAT	23
	SMALLFLT	23
Character string data	CHARACTER VARCHAR	Defined length	If fixtext_option is specified in the enclose operand, add 2 to the output length.
Mixed character string data	MCHAR MVARCHAR	Defined length
National character data	NCHAR NVARCHAR	Defined length 2
Date data	DATE	10
Time data	TIME	8
Date interval data	INTERVAL YEAR TO DAY	10
Time interval data	INTERVAL HOUR TO SECOUND	8
Time stamp data	TIMESTAMP	Decimal part 0:19 2:22 4:24 6:26
Large object data	BLOB	0
Binary data	BINARY	0
Abstract data type	ADT	0

⁵ If you are reorganizing files in units of schemas (including unload files), use the sum of the values obtained for individual tables.