Performance Related Engine Properties Settings

When true events go directly to the SLC and FileAgent services, which is more efficient, instead of indirectly via the EventMonitor service.

May only be set to true for single EP installations.

Introduced in v6.1.2.1, this property communicates to the CEP to reset the status of all monitored servers to Unknown at startup instead of leaving them in their current state, which could cause invalid status information to be shown until all EPs have been started and begun monitoring servers they are the active EP for.

Used to limit the date range used by the Web console queries for data to less (or potentially more) than a years’ worth of information, which can result in far better performance.

Set this property lower for better Web console performance, albeit causing it to show less data.

Informs IBM® Sterling Control Center Monitor whether at startup, and subsequently once a day, to initiate queries to ascertain the row counts in various database tables it stores information in. The values obtained are only seen in the IBM Sterling Control Center Monitor Web console.

These queries can be onerous to run - they can take a long time and cause database contention. Set this property to TRUE for best performance.

In v6.1.2.1 and above, changes made to this property do not require IBM Sterling Control Center Monitor to be restarted to take effect.

Tells the EventMonitor the count of events that will trigger it to enter catchup mode, which means it will not wait the normal amount of time before attempting to retrieve more event data for processing, which is subsequently passed on to the SLC and FileAgent services.

Any changes made to this property do not require IBM Sterling Control Center Monitor to be restarted to take effect.

Communicates to EventMonitor the query limit for retrieving event data.

The default value for this property is evolving and for best performance the current recommendation is to set it to 1000.

In v6.1.2.1 and above, any changes made to this property do not require IBM Sterling Control Center Monitor to be restarted to take effect.

When this property, available in v6.1.2.1 and above, is set to TRUE the EventMonitor logic limits the query used for retrieving event data to only get data the SLC service requires, which makes the query, and the EventMonitor, run faster.

For best performance set this property TRUE but you should not do this if Connect:Direct File Agents are being monitored.

Changes made to this property do not require ICC to be restarted to take effect.

Introduced in v6.1.2.1, this property informs the EventMonitor logic how many threads to use while reconstructing IBM Sterling Control Center Monitor events from database data.

Changes made to this property do not require IBM Sterling Control Center Monitor to be restarted to take effect.

Default: Number of CPUs the server running IBM Sterling Control Center Monitor has.

Introduced in v6.1.2.1, this property informs how many seconds to wait in-between checks for status of Connect:Direct File agents.

When monitoring 1000s of File agents this number should be increased to limit the amount of time spent status checking by the FileAgentService.

Changes made to this property do not require IBM Sterling Control Center Monitor to be restarted to take effect.

Introduced in v6.1.2.1, this value limits the number of ready to be summarized processes that will be queried for by the ProcessSummaryService.

This property exposes a previously hard-coded value that was already in the logic.

Changes made to this property do not require IBM Sterling Control Center Monitor to be restarted to take effect.

Introduced in v6.1.2.1, this value in milliseconds communicates to IBM Sterling Control Center Monitor when it should generate warnings in the IBM Sterling Control Center Monitor log files about slow processing of SNMP traps received from Connect:Direct File agents.

Prior to v6.1.2.1, server reassignments were performed one at a time that is, serially. In v6.1.2.1, multiple threads may be used to speed up the server reassignment process.

This property is used in conjunction with the properties SERVER_REASIGNMENT_QUEUE_SIZE and SERVER_REASSIGNMENT_THREAD

For best server reassignment performance this property should be TRUE.

Changes made to this property do not require ICC to be restarted to take effect.

Prior to v6.1.2.1, NodeServices for Connect:Direct servers would stay connected, if possible, between polls of their respective servers.

While it is less efficient to disconnect between polls, to facilitate monitoring of thousands of servers without running out of server socket connections, you can now configure IBM Sterling Control Center Monitor to do so, which limits the number of socket connections ICC keeps open at one time.

Changes made to this property do not require IBM Sterling Control Center Monitor to be restarted to take effect.

Sets the limit used in queries initiated by the queueUnprocessedEvents thread to retrieve unprocessed OSA events.

This property is only used when OSA event processing is done asynchronously, which is controlled by the MULTI_THREAD_OSA_EVENT_PROCESSING property.

Sets the limit on the queue used to contain unprocessed OSA events retrieved by queueUnprocessedEvents thread, which are subsequently processed by the handleUnprocessedEvents thread.

Always make the value for this property be some multiple, greater than one, of the OSA_EVENTS_TO_GET_AT_ONCE_COUNT value.

This property is only used when OSA event processing is done asynchronously, which is controlled by the MULTI_THREAD_OSA_EVENT_PROCESSING property.

Dictates how many threads are used by the ProcessSummarizationService to summarize completed processes.

The larger this value is set; the higher memory and CPU utilization will become.

Dictates how many processes may be summarized at once by each process summarization thread, which are called Process Summary Workers.

The larger this value is set; the higher memory utilization will become.

In v6.1.2.1 and above, changes made to this property do not require IBM Control Center to be restarted to take effect.

Controls the frequency at which the QueuedProcessesClearJob runs, which ensures completed B2Bi processes are removed from the Queued process view.

The larger this value is made the less contention there will be on the CC_PROCESS database table, and the longer completed B2Bi processes may remain in the Web console queued process view.

Set this property value higher for better performance.

Added in v6.1.2.1, this property is count of server reassignments that may be queued for handling.

Note this property is used in conjunction with the property MULTI_THREAD_SERVER_REASSIGNMENT.

Prior to v6.1.2.1, server reassignments were handled synchronously that is, one at a time.

Multiple threads may be configured as of v6.1.2.1 to handle reassignments and will be by default.

Prior to v6.1.2.1 NodeService threads for monitor servers were started one at a time when EPs started, and they were stopped one at a time when EPs were shutdown.

As of v6.1.2.1, multiple threads may be used to speed up the startup and shutdown processes.

This property is used in conjunction with the property USE_MULTIPLE_THREADS_FOR_NODESERVICE_STARTUP_AND_SHUTDOWN.

This property value, a value in milliseconds, limits the query used by the ProcessSummaryService when searching for processes ready to be summarized. Processes that completed less than the amount of time specified will not be returned.

This is not a new property, but the default was changed from 0 to 1000 in v6.1.2.1 to avoid occasional problems detected during testing that caused events to not be associated with completed processes when they should have been.

In v6.1.2.1, changes made to this property do not require IBM Sterling Control Center Monitor to be restarted to take effect.

Used as a row limit in the SQL used to delete summary table data.

Since the SQL used to delete data causes more contention on the database than the SQL used to find the data to be deleted, the value for this property should be kept relatively small compared to the property SUMMARY_TABLES_PURGE_QUEUE_SIZE.

Communicates to IBM Sterling Control Center Monitor to also purge data from the FILE_COUNTS database table when other summary data is purged.

For best performance set this property to TRUE.

Communicates to IBM Sterling Control Center Monitor to also purge data from the ROLL_UP database table when other summary data is purged.

For best performance set this property to TRUE.

A value in milliseconds that dictates how long the queueUnprocessedEvents thread waits after retrieving and queuing unprocessed OSA events.

Note when the number of OSA events previously retrieved matches the OSA_EVENTS_TO_GET_AT_ONCE_COUNT property value no wait is performed. This is referred to as catchup mode.

This property is only used when OSA event processing is done asynchronously, which is controlled by the MULTI_THREAD_OSA_EVENT_PROCESSING property.

A value in milliseconds that dictates how long the handleUnprocessedEvents thread waits after processing the previous batch of OSA events before attempting to process more.

Note when the batch size of OSA events previously processed matches the OSA_EVENTS_TO_PROCESS_QUEUE_SIZE size, no wait occurs. This is referred to as catchup mode.

This property is only used when OSA event processing is done asynchronously, which is controlled by the MULTI_THREAD_OSA_EVENT_PROCESSING property.

Prior to v6.1.2.1 the time to add servers grew by a significant amount as the number of monitored servers increased. Inefficiencies found in the logic used to update server groups were primarily to blame for the slowness. The logic was updated to make updates of server groups significantly faster and this property controls whether the new logic is used.

For best performance this property should be TRUE.

Changes made to this property, in v6.1.2.1, do not require IBM Sterling Control Center Monitor to be restarted to take effect.

Prior to v6.1.2.1 NodeService threads, that monitor servers, were started one at a time when EPs started, and stopped one at a time when EPs were shutdown. Now multiple threads may be used to speed up the IBM Sterling Control Center Monitor startup and shutdown processes.

This property, introduced in v6.1.2.1, is used in conjunction with SERVER_STARTUP_SHUTDOWN_THREADS.

For best performance this property should be TRUE.

SUMMARY_TABLE_PURGE_FILE_COUNTS.

Tells IBM Sterling Control Center Monitor to also purge data from the FILE_COUNTS database table when other summary data is purged. For best performance set this property to TRUE.