Page tree
Skip to end of metadata
Go to start of metadata


Avg user cmd exec time for <user_task_ID>: <duration> ms, <number_of> commands, <appl_pid_and_node>



The time needed by the specified user tasks to execute the statements is very long. You see the average execution time of a statement, the number of statements checked, and the process ID of the relevant application process.

Whether this counts as a bottleneck depends on the application structure.

Mass statements in background processes can often cause long runtimes. As well as this, situations such as locks in SQL objects, physical reads and writes, or dispatching caused by the prioritization of other tasks, can cause internal kernel wait situations that increase runtimes.

If, in SAP liveCache databases, a DB procedure takes a long time to execute, it is highly likely that other users are being blocked in the same SAP liveCache thread. This can cause unexpectedly long runtimes for other transactions or DB procedures.


User Response

Find out if this is processing of mass data with justified long runtime or if the command runtime can be improved f.e. by definition of an index.


More Information

Relevant  DB-Analyzer Files
Helpful with ...
DBAN_RUNNING_COMMANDS.prt Find out which commands were running at the time of the DB-Analyzer snapshot. 
SAP Note
Expert Sessions (Link to trainings)
Content / Helpful with ...

Session 16: SAP MaxDB SQL Query Optimization (Part 1) Session 16: SAP MaxDB SQL Query Optimization (Part 2)

Long running commands may be caused by disadvantegeous execution plans. Find out how the optimizer works to create adequate indexes. 
Session 4: Performance Optimization with SAP MaxDB  Find information about the command and resource monitor and general aspects of performance analysis. (Parts ot the information may be outdated but the general principles should last.) 
Session 27: SAP MaxDB Multi TaskingProvides general information about the concept of "cooperative" multi-tasking.
Documentation (Links)
Helpful with ...


  • No labels