Workload Monitoring

Workload monitoring in IT-Conductor represents the health status in a graphic and detailed view of the system in terms of usage to monitor response times in transactions, users, expensive running programs, etc.

Workload represents the usage of system resources by units to complete business processes such as the following:

• Transactions

• Jobs

• Programs

• Sessions or Tasks

Why Monitor Workloads?

System administrators must always be aware of the system's health status to prevent crashes or unexpected incidents that can impact the business.

The three points that we will analyze are the following:

• Performance: Workloads to measure how much work can be done and to monitor response times and processing windows to guarantee the operativity of the services.

• Capacity & Configuration: Provisioning of capacity in an early stage to prevent any issue.

• Baseline & Trend: Capture the performance statistics.

What’s Available for Workload Monitoring in SAP?

• SAP CCMS Transaction SLA: Response time per app/server/client, smoothed out per interval in RZ20.

• Netweaver Workload Analysis: ST03 – Analyzes SAP statistical records from a historical reporting perspective, detailed transaction records but without context.

• Solution Manager: IT Performance Report, EarlyWatch Report: historical snapshots of workload types and top workloads in an aggregated format. Not monitoring.

Why Our Approach to Workload Monitoring is Beneficial?

The workload retriever filters matching criteria and brings detailed statistics for analytics, providing in-depth performance intelligence at the OS, database, and application levels.

Service-Oriented Workload Monitoring

• Service Alert: Set customized alerts with value ranges defined by the business to determine the health status of the systems (OS, APP, DB).

Figure 1: Sample Service Alert Graph

• Overrides: Create Overrides to set customized metrics and severity ranges to monitor the system.

Figure 2: Sample System Overrides

• System Availability: Set metrics to assure that the system is working as required.

Figure 3: Sample Availability Graph

• Memory Used: Set metrics and alerts when memory exceeds the warning value previously configured in overrides.

Figure 4: Sample Memory Used Graph

• Dialog Response Time: Set metrics to monitor the average time required to process dialog steps.

Figure 5: Sample Dialog Response Time Graph

• Dialog Users Logged in: Review how many users have logged into the system in a period of time.

Figure 6: Sample Dialog Users Logged In Graph

• Active Dialog Users: Review how many users connected from GUI in a period of time.

Figure 7: Sample Active Dialog Users Graph

CCMS Alerts

• Details for Failed Batch Jobs: Get information about failed batch jobs in a period of time.

Figure 8: Sample Failed Batch Jobs

• Details for Failed Updates: Get information about failed updates in the system in a period of time.

Figure 9: Sample Failed Updates

• Details for Shortdumps: Get information about short dumps in the system in a period of time.

Figure 10: Sample Shortdumps

Service-Oriented Workload Monitoring helps SAP Basis teams to answer the following questions:

• What was running on the system during that time?

• Who was running that problem transaction or job?

• Where did those problem users or transactions come in from?

• When did the performance problem start?

• How was the workload or services impacted?

Video