A TASM exception can detect when a request is skewed. However, you need to create a system event to detect a system-wide skew, such as an application running on one node. The system event CPU Skew can detect any type of skew, including session and application imbalances. Typically, when TASM detects a system skew, it sends an alert to the DBA, who investigates and acts manually.
Teradata recommends that you use the Averaging qualification option to distinguish between a temporary skew and a persistent skew.
If you are using CPU Skew on a coexistence system, adjust the triggering threshold to a value appropriate for the built-in system imbalance. For example, suppose in a perfectly balanced workload environment, the typical utilization of 10 old nodes is 95% when 10 new nodes are exhausted at 100%. Here, the built-in system skew is (100 - 95) / 100 = 5%. In this case, set the CPU Skew Skew Percent to a value greater than 5%.