ITT - Integrated Task Time
`Integrated Task Time (ITT)` is a cumulative measure of the total processor time consumed by a task or address space in the z/OS environment. It includes both `Task Control Block (TCB)` time and `Service Request Block (SRB)` time, providing a comprehensive view of the CPU resources directly attributable to a specific workload. ITT is a fundamental metric for performance analysis, accounting, and chargeback in mainframe systems.
Key Characteristics
-
- Comprehensive CPU Metric:
ITTaggregatesTCBtime (CPU time spent executing user code and system services on behalf of the task) andSRBtime (CPU time spent by system routines performing services for the task, often asynchronously). - Accounting and Billing Basis: It is a primary metric used by z/OS for workload accounting, chargeback, and billing purposes, as it represents the direct CPU consumption of a job or started task.
- Granularity:
ITTis typically measured in microseconds or milliseconds and is recorded by the system for each task, step within a job, or for an entire address space. - Excludes I/O Wait Time:
ITTspecifically measures CPU execution time and does not include time spent waiting for I/O operations to complete, making it a pure measure of processor utilization. - Recorded in SMF:
ITTvalues are meticulously recorded inSystem Management Facilities (SMF)records (e.g., Type 30 records for jobs/steps, Type 72 for WLM reporting), making it available for post-processing and analysis. - Reflects Workload Demand: High
ITTvalues indicate significant CPU demand from a particular workload, which is crucial for performance tuning and capacity planning.
- Comprehensive CPU Metric:
Use Cases
-
- Workload Performance Analysis: Analyzing
ITTfor specific jobs, started tasks, or applications helps identify CPU-intensive workloads that might be candidates for optimization or require additional processing capacity. - Chargeback and Billing: Many mainframe installations use
ITTas a primary metric to charge internal departments or external clients for their CPU resource consumption. - Capacity Planning: Tracking
ITTtrends over time for critical applications assists in forecasting future CPU requirements and planning for hardware upgrades or workload rebalancing. - Problem Determination: A sudden or unexpected increase in
ITTfor a stable application can signal a performance regression, an inefficient code change, or an unexpected workload spike, prompting investigation. - Workload Manager (WLM) Goal Setting: While WLM primarily manages based on service units, understanding the
ITTconsumption of a workload helps in defining appropriate WLM service classes and performance goals.
- Workload Performance Analysis: Analyzing
Related Concepts
ITT is a critical component of overall CPU utilization on z/OS, directly combining TCB time and SRB time, which are the two primary categories of CPU consumption for a task. It is distinct from elapsed time, as ITT only measures active CPU work, excluding I/O wait time. ITT values are captured in SMF records, which are then processed by tools like RMF (Resource Measurement Facility) or third-party performance monitors to provide insights into system and workload performance. WLM uses ITT (converted to service units) as part of its resource management decisions, aiming to