Adaptive Server provides resource limits to help system administrators prevent queries and transactions from monopolizing server resources. Resource limits, however, are not fully specified until they are bound to a time-range.
A resource limit is a set of parameters specified by a system administrator to prevent an individual login or application from exceeeding the limits described in Table 1-1.
LIimit being tracked |
What is tracked by Adaptive Server |
|---|---|
Exceeding I/O costs |
Estimated costs – as determined by the optimizer. Actual costs – as measured during query execution |
Returning excessive number of rows |
Tracked on a per-query basis |
Exceeding a given elapsed time |
A query batch basis in a given transaction. |
Utilizing excessive tempdb space |
Tracked per session |
Idle time |
Tracked per process |
Resource limits are bound to time-ranges, allowing the system administrator to define precisely when they should be enforced. When the system administrator modifies a resource limit, all users logged in the session see the change, including the system administrator
The set of parameters for a resource limit includes the time of day to enforce the limit and the type of action to take. For example, you can prevent huge reports from running during critical times of the day, or kill a session whose query produces unwanted Cartesian products.
