17.10 - Active Requests - Advanced SQL Engine - Teradata Workload Management

Teradata Vantageā„¢ - Workload Management User Guide

Advanced SQL Engine
Teradata Workload Management
Release Number
Release Date
July 2021
Content Type
User Guide
Publication ID
English (United States)

The Active Requests event lets you monitor how many requests are running in a workload. Active Requests does not include requests held in the delay queue.

If the number of active requests stays high, that indicates either unmanaged arrival rate surges and lulls or other unusual situations. Too many active requests can cause the following problems:
  • The exhaustion of critical shared resources such as AWTs, memory, and physical spool
  • The possibility of entering flow control or congestion management

One valuable use of the Active Requests event is to detect when too many requests are running in a penalty-box workload. Typically, a request enters a penalty-box workload because it is running too long and triggers an exception. Requests that are already running cannot be throttled, so you cannot use throttles to control requests in the penalty-box workload. If a request in the penalty-box workload holds any critical, shared resources, it will probably hold those resources for a long time because of the low priority given to the penalty-box. This means that when the penalty-box has many active requests, there is a strong chance that higher-priority work is affected. If more than 3 or 4 requests are running in the penalty-box workload, the Active Requests event action can notify the DBA so that he can abort some requests manually.