The following overview describes the special database parameters. For more information about other database parameters, see General Database Parameters, Support Database Parameters and liveCache Database Parameters.
Changes to database parameters do not take effect until the database instance is restarted.
With the Database Manager CLI, you can change some database parameters even while the system is running.
Special Database Parameters (Extended)
Parameter |
Meaning |
How can I change this Parameter? |
_COMMENT |
Description of the database instance |
Restart required after change |
_DATA_CACHE_RGNS |
Number of critical sections into which the data cache is divided This division enables parallel writing to the data cache. The parameter is calculated by the database system and cannot be changed. You can override the parameter for testing purposes, however, using the XP_DATA_CACHE_RGNS support database parameter. |
Parameter cannot be changed |
_MAX_MESSAGE_FILES |
Maximum number of trace files that can be open simultaneously |
Restart required after change |
_MAXEVENTS |
Maximum number of events stored in the working memory by the kernel that are to be processed by the Database Manager |
Restart required after change |
_ROW_RGNS |
Number of critical sections into which the total of all rows are divided in the lock list This division enables parallel access to the lock list. |
Restart required after change |
_TAB_RGNS |
Number of critical sections into which the total of all tables in the lock list are divided This division enables parallel access to the lock list. |
Restart required after change |
_TRANS_RGNS |
Number of critical sections into which the total of all transactions in the lock list are divided This division enables parallel access to the lock list. |
Restart required after change |
_UNICODE |
Defines whether the database system uses UNICODE as the character set |
You define this database parameter when you create the database instance. Afterwards you can only change the database parameter by re-initializing the database instance. However, in doing so you lose all application data. |
AUTO_RECREATE_BAD_INDEXES |
Indexes marked as BAD are recreated automatically after a restore. |
Restart required after change |
BACKUP_BLOCK_CNT |
Block size used by the database system when writing to backup media Only useful for tapes that have an optimal block size |
Restart required after change |
CAT_CACHE_SUPPLY |
Memory size of the catalog cache for all user tasks (in pages) |
Restart required after change |
CONVERTER_REGIONS |
Number of critical sections into which the converter is divided This division enables parallel writing to the converter. The parameter is calculated by the database system and cannot be changed. You can override the parameter for testing purposes, however, using the XP_CONVERTER_REGIONS support database parameter. |
Restart required after change |
DATE_TIME_FORMAT |
System default for the date and time format Each user can define the date and time format for a session by making corresponding entries in the database tool (such as in the Loader using standard_date_mask) or by making corresponding entries in SQL statements (datetimeformat). See Date and Time Format. |
Parameter can be changed with the Database Manager CLI while in operation |
DEADLOCK_DETECTION |
Maximum search level for deadlock detection Any deadlocks that the database system does not detect at this search level are only resolved by the parameter REQUEST_TIMEOUT. |
Restart required after change |
DEFAULT_CODE |
System default for the code attribute, applies only to the column values of the data types CHAR[ACTER], VARCHAR and LONG[VARCHAR]. |
Parameter can be changed with the Database Manager CLI while in operation |
DIAG_HISTORY_NUM |
Number of histories in the <diag_history_path> directory |
Restart required after change |
DIAG_HISTORY_PATH |
Directory <diag_history_path> in which the database system saves diagnosis files (backups of the most important log files and memory dumps) after a database error |
Restart required after change |
FILEDIR_SPINLOCKPOOL_SIZE |
Number of spinlocks available for directory reader and writer lock operations |
Restart required after change |
FORMAT_DATAVOLUME |
Suppresses the formatting of data volumes when a new database instance is created; only useful for test purposes |
Restart required after change |
FORMATTING_MODE |
The database system formats the volumes when the database instance is created. With this parameter, you define the type of formatting. Parallel formatting is only useful for data volumes of the file type which are located on different hard disks. |
Restart required after change |
HASHED_RESULTSET |
Specifies whether the database system, using a hash procedure, optimizes result sets that result from the execution of a join |
Parameter can be changed with the Database Manager CLI while in operation |
HASHED_RESULTSET_CACHESIZE |
Maximum size of the working memory that the database can use for result sets that are optimized with a hash procedure (in KB) |
Parameter can be changed with the Database Manager CLI while in operation |
HIRES_TIMER_TYPE |
Type of time measurement that is used by the kernel for internal operations; only useful for multiprocessor computers on which there can be synchronization problems between the CPUs |
Restart required after change |
HS_DELAY_TIME_<NNN> |
Only for hot standby systems Delay in seconds until the standby instance NNN redoes the changes in the master instance |
Restart required after change |
HS_NODE_<NNN> |
Only for hot standby systems Computer name / IP address of the hot standby instance NNN The default value for the master instance is HSS_NODE_001. |
Restart required after change |
HS_STORAGE_DLL |
Only for hot standby systems Name of the library through which the hot standby system addresses the memory management system The default value is libhsscopy. |
Restart required after change |
HS_SYNC_INTERVAL |
Only for hot standby systems Minimum time span between two commands of the master instance to synchronize the standby instances |
Restart required after change |
JOIN_MAXTAB_LEVEL4 |
Parameter of the SQL Optimizer; see JOIN_SEARCH_LEVEL |
Parameter can be changed with the Database Manager CLI while in operation |
JOIN_MAXTAB_LEVEL9 |
Parameter of the SQL Optimizer; see JOIN_SEARCH_LEVEL |
Parameter can be changed with the Database Manager CLI while in operation |
JOIN_SEARCH_LEVEL |
Algorithm used by the SQL Optimizer for the join sequence search; see Database Parameter JOIN_SEARCH_LEVEL |
Parameter can be changed with the Database Manager CLI while in operation |
JOIN_TABLEBUFFER |
Size of the buffer that the SQL Optimizer can use for each join step for a read operation |
Restart required after change |
KERNELDIAGSIZE |
Size of the log file knldiag (in KB) |
Restart required after change |
LOAD_BALANCING_CHK |
Time interval in which the database system checks whether load balancing is useful (in s) |
Restart required after change |
LOG_BACKUP_TO_PIPE |
Specifies whether a log backup in pipes is allowed NO: After a log backup has been exported to a pipe, the connection to the pipe is then closed without the database receiving any information on whether the log backup was successful. As a result, the log area is not released and cannot be overwritten, in which case it may become full and further data changes impossible. YES: Note that in the case of a failed log backup, the corresponding log entries in the log area may already have been overwritten, in which case you would not be able to restore the database completely after a crash. |
Restart required after change |
LOG_IO_QUEUE |
Size of the log queue (in pages) |
Restart required after change |
LOG_QUEUE_COUNT |
Number of log queues =0: Number calculated from the MAXCPU parameter >0: Value of the USED_LOG-QUEUE_COUNT parameter is adopted |
Restart required after change |
LRU_FOR_SCAN |
Specifies whether the whole data cache is used for scans |
Parameter can be changed with the Database Manager CLI while in operation |
MAX_HASHTABLE_MEMORY |
Maximum size of the working memory that the database system can use to cache all tables when executing joins (in KB) |
Parameter can be changed with the Database Manager CLI while in operation |
MAX_SINGLE_HASHTABLE_SIZE |
Maximum size of the working memory that the database system can use to cache a single table when executing a join (in KB) |
Parameter can be changed with the Database Manager CLI while in operation |
MAXRGN_REQUEST |
Maximum number of attempts a task should make to access a region If this number is exceeded, the task releases the access to the CPU for a different task of the same user kernel thread. |
Parameter can be changed with the Database Manager CLI while in operation |
MAXSERVERTASKS |
Number of server tasks |
Restart required after change |
MEM_ENHANCE_LIMIT |
Maximum additional memory to be used, which is possible through a Memory Scalability Enhancement for example (in MB) |
Restart required after change |
MEMORY_ALLOCATION_LIMIT |
Maximum memory to be used by the database instance (in KB) |
Restart required after change |
MP_RGN_LOOP |
Maximum number of times a task may attempt to access a critical section that has been locked by another task If this number is exceeded, the accessing task changes its state to “Waiting”. |
Parameter can be changed with the Database Manager CLI while in operation |
OFFICIAL_NODE |
Only for hot standby systems Virtual server name by which the cluster is addressed from outside The system administrator initially assigns the virtual server name to the computer on which the master instance of the hot standby system is located. If the master instance breaks down, then the name transfers to the standby instance, which takes on the master role. |
Restart required after change |
OPTIM_INV_ONLY |
Defines whether the SQL Optimizer uses the index-only strategy for joins |
Parameter can be changed with the Database Manager CLI while in operation |
OPTIM_MAX_MERGE |
Influences the decision as to whether the SQL Optimizer uses an index-merge strategy If the number of pages of an index that need to be merged exceeds the value specified in OPTIM_MAX_MERGE, the SQL Optimizer does not use this index for an index-merge strategy. |
Parameter can be changed with the Database Manager CLI while in operation |
OPTIMIZE_QUERYREWRITE |
Automatic Transformation of SQL Statements Default value for software versions >= 7.6: OPERATOR (the database system transforms SQL statements). To activate the old transformation behavior (versions < 7.6) choose STATEMENT. |
Parameter can be changed with the Database Manager CLI while in operation |
REQUEST_TIMEOUT |
Maximum wait for a lock to be released (in s); for all database sessions, limits wait times for a lock to be released by another user. If a lock request cannot be satisfied within the time defined here, a message is sent to the waiting database session. The database system rolls back any changes previously executed within the transactions (ROLLBACK). |
Restart required after change |
SEQUENCE_CACHE |
Size of the sequence cache (in pages) |
Restart required after change |
SESSION_TIMEOUT |
Timeout value for database sessions (in s) |
Parameter can be changed with the Database Manager CLI while in operation |
SHAREDSQL |
Activating Shared SQL |
Parameter can be changed with the Database Manager CLI while in operation |
SHAREDSQL_COMMANDCACHESIZE |
Maximum size of the Shared SQL cache (in KB) |
Restart required after change |
SHAREDSQL_EXPECTEDSTATEMENTCOUNT |
Expected number of SQL statements to be stored in the shared SQL cache |
Restart required after change |
TRACE_PAGES_<task> |
Maximum number of pages that are available in the individual task types for each user kernel thread Each user kernel thread has its own trace area in the main memory. <task> can have the following values: BUP (Backup Task, not yet implemented), EV (Event Task), GC (Garbage Collector), LW (Log Writer), PG (Pager), SV (Server Task), TI (Timer Task), US (User Task), UT (Utility Task) |
Restart required after change |
USED_LOG_QUEUE_COUNT |
See LOG_QUEUE_COUNT |
Restart required after change |
VOLUMENO_BIT_COUNT |
Number of bits in the converter block address that are reserved for the logical volume number of a data volume Permitted values: 6 ≤ VOLUMENO_BIT_COUNT ≤ 12 (see Technical Restrictions) We recommend that you do not change the value that was selected when the database instance was created. If you change the database parameter when the database is running, you then have to restore the database instance to update the data volume numbering. |
Restart required after change |
XP_CONVERTER_REGIONS |
Overriding the parameter CONVERTER_REGIONS |
Restart required after change |
XP_DATA_CACHE_RGNS |
Overriding the parameter _DATA_CACHE_RGNS |
Restart required after change |
XP_MAXPAGER |
Overriding the parameter MAXPAGER |
Restart required after change |