SQL/MX 3.2.1 Management Manual (H06.26+, J06.15+)

INSERT, UPDATE, or DELETE operation to change a row in a table. If the change would violate
the constraint, NonStop SQL/MX prohibits it.
Concurrency control for database access provides different degrees of database consistency to
meet different needs. For any DML statement, you can specify access options by using the FOR
option ACCESS clause and, for a SELECT statement, by using this same clause, you can also
specify access options for individual tables referenced in the FROM clause. The possible settings
for options in a DML statement are: READ COMMITTED, READ UNCOMMITTED, SERIALIZABLE
or REPEATABLE read, SKIP CONFLICT, and STABLE. These access modes are maintained by shared
or exclusive locks on rows, sets of rows, partitions, and tables. For more information, see the
SQL/MX Reference Manual.
Multiple Character Sets
NonStop SQL/MX allows you to associate one of these character sets with a literal or host variable:
Default single-byte 8-bit character set for character data types. It supports English and other Western
European languages.
ISO88591
Double-byte Unicode character set in UTF16 big-endian encoding. All Basic Multilingual Plan (BMP)
characters are included. Surrogate characters are treated as two double-byte characters.
USC2
For more information about supported character sets, see the SQL/MX Reference Manual. For
more information about the supported character sets for NonStop SQL/MP, see the SQL/MP
Reference Manual.
Database Security
Authorization to operate on SQL/MX tables, views, and stored procedures in Java (SPJs) is
maintained by the GRANT and REVOKE statements.
SQL/MX programs and files stored in OSS space are subject to the same security rules as any
other files stored there.
For more information, see “Planning Database Security and Recovery” (page 42).
Parallel Processing
NonStop SQL/MX performs these types of parallel processing:
• Parallel query processing
• Parallel join operations
• Parallel index maintenance
• Parallel index loading
• Parallel sorting
• Parallel input-output operations
Parallel query processing can provide speed-up and scale-up. Speed-up is the reduction of response
time, which you can accomplish by spreading the database over multiple partitions. Scale-up is
the maintenance of a constant response time even though the database is growing, which you can
accomplish by increasing the number of partitions.
When tables and indexes are partitioned across multiple disks, NonStop SQL/MX can use a
different process for each partition during query execution. This approach reduces the time for
scans and other set-oriented processing by a factor equivalent to the number of partitions when
there is no contention in the CPU-controller-disk path; that is, when every participating disk is
primary to a different CPU.
Parallel join operations are performed by the SQL/MX executor during query processing.
SQL/MX Database Features 21