SQL/MX Release 2.0 Best Practices

Data-loading Techniques and Considerations 38
for data load are IMPORT, the DataLoader/MX programming API, custom programming, or third-party ETL
tools.
Getting Guardian Names from the System Catalog for FUP RELOAD
Note that definition_schema_version_1200 is correct for now; however, in the future, there will be version
numbers higher than 1200.
set schema definition_schema_version_1200;
select 'fup reload ', substring(p.system_name, 1, 15) ,
'.' , rtrim(p.data_source) ,
'.' , rtrim(p.file_suffix) ,
', rate 40'
from partitions p, objects o
where o.object_uid = p.object_uid
and o.schema_uid =
(select schema_uid
From nonstop_sqlmx_SYSTEMNAME.system_schema.schemata
Where schema_name = 'SCHEMANAME'
And cat_uid =
(select cat_uid
From nonstop_sqlmx_SYSTEMNAME.system_schema.catsys
Where cat_name = 'CATALOGNAME'
For read uncommitted access
)
For read uncommitted access
)
and o.object_name not like 'HISTOGRAM%'
and o.object_name not like 'MVS%'
and o.object_name_space in ('IX','TA')
order by p.system_name, p.data_source, p.file_suffix
for read uncommitted access;
Note: schemaname and catalogname in the query are internal format ANSI identifiers. Regular identifiers
must be specified in all uppercase. Delimited identifiers must be specified without surrounding double-
quotes.
Block Slack Space
Scan performance for database queries depends largely on data compaction—that is, how much data
a single I/O request will return. This, in turn, depends on the degree of table fragmentation, internal block
free space, and row size. The more compact and organized the data blocks are, the better the scan
performance will be.
One way to ensure that data blocks are as compact as possible is to reload the table data, using Slack
to 0. SQL/MX software uses a default Slack parameter of 15 percent, meaning that 15 percent of the