Application Performance Monitoring
and Analysis

Compuware Strobe is the industry-leading mainframe application performance management and analysis solution for mainframe applications. With Strobe, IT departments can pinpoint application inefficiencies that cause excessive CPU consumption, elongated elapsed times and reduce hardware and software costs while increasing customer satisfaction.

Key Features and Benefits of Mainframe Application Performance Management

  • Improve user code efficiency, transaction response time and throughput
  • Locate and improve resource-consumptive database calls
  • Quickly pinpoint the source of application inefficiencies
  • Identify SQL statements and DB2 system services that cause excessive CPU time or wait time
  • Receive detailed DB2 tuning information including catalog statistics and the number of GET PAGE requests for each SQL statement
  • Enable new users to be more productive with a graphical interface for taking measurements and detecting mainframe application performance problems
  • Quickly and easily identify tuning opportunities that will have the greatest impact on monthly software licensing costs by launching Strobe directly from BMC’s MainView or Cost Analyzer for zEnterprise
  • Resolve emerging performance issues before they impact application service levels by automatically launching Strobe directly from ConicIT as soon as anomalies/potential issues are detected
  • Gain intelligence on individual transactions as they flow from an end-user’s device through various layers of servers and back to the device through Strobe’s integration with tools from AppDynamics and Dynatrace

Reduce the Complexity of Mainframe Application Performance Management

Strobe is the premier technology for managing mainframe application performance in complex environments. It measures the activity of both online and batch-processing applications in single and multiple system environments, detailing where and how time is spent during an application’s execution.

Strobe can save hours of investigation by revealing program statements that consume excessive CPU resources, the percentage of CPU time used within modules and control sections, display procedure names and statement numbers, and provide the complete line of source code. With Strobe analysis you can identify a number of I/O problems including inefficient VSAM file buffering.

Compuware iStrobe extends the benefits of Strobe with an easy-to-use, browser-based graphical user interface. iStrobe identifies the access path used by the SQL statement as well as the relevant catalog statistics, providing all the data needed to improve the application’s performance for both online and batch-processing applications. Additionally, iStrobe allows the user to perform “What If” analysis by making changes to SQL to see how they might affect the DB2 access path.

Many applications today span multiple CICS regions in a CICSPlex. iStrobe’s CICSPlex reporting aggregates the data from multiple CICS measurements into a single report. Similarly for IMS, IMS transactions may execute in many IMS Message Processing Regions (MPRs) within an IMS subsystem. iStrobe will capture data on all the transactions providing the “single pane of glass” for analyzing the performance of an IMS transaction.

Proactively Determine What and When to Measure

Compuware AutoStrobe boosts productivity and increases the flexibility of Strobe and iStrobe by making the measurement process predictable and repeatable. You can create reusable measurement requests and schedule them to run at regular intervals. In addition, AutoStrobe can proactively initiate a Strobe measurement when abnormal behavior occurs in a program. Third-party monitors or other custom applications can also initiate a Strobe measurement using the Strobe applications programming interface (API).

AutoStrobe also includes Global Batch Monitoring, a feature that automates the detection of performance problems by using a rolling average of CPU consumption and elapsed time metrics. AutoStrobe automatically measures any job step that exceeds acceptable thresholds, while Global Batch Monitoring automatically sets CPU and elapsed time thresholds based on performance data from previous executions. A similar function for CICS online transactions monitors transaction performance and alert you to any performance issues before they become critical.

Integrations with Industry-leading tools

Mainframe workloads are growing significantly as organizations embrace mobile apps, Big Data analytics, IoT and other technologies. These technologies all place greater demands on mainframe-resident applications, data and transaction/compute processing power. New capabilities offered on the newly debuted IBM z13—including Java-enabled specialty processors—are also likely to attract new workloads. This workload growth drives up the Monthly License Charge (MLC).

To help customers minimize their MLCs, Strobe is tightly integrated with industry-leading products from BMC. The integration between these solutions enables customers to quickly and easily identify tuning opportunities that will have the greatest impact on their monthly software licensing costs by applying BMC Cost Analyzer’s visual mapping to the detailed batch and transaction information provided by Compuware Strobe. These opportunities can include moving workloads to non-peak periods, running IBM subsystems on fewer LPARs, and capping LPAR utilization. The integration also makes cost management easier for mainframe operations staff by allowing them to invoke Strobe data collection directly from BMC MainView. Further information can be found here.

Strobe also integrates with ConicIT to help Ops staff proactively resolve emerging mainframe issues before they impact application service levels. ConicIT baselines normal mainframe behavior across thousands of metrics and then applies advanced algorithms to recognize in real time when those metrics deviate from the baseline. Predictive alerts from ConicIT automatically launch a Strobe measurement to provide the root-cause analysis to address the appropriate corresponding mainframe conditions. Strobe empowers even mainframe-inexperienced systems staff to act on those alerts by guiding them through an exploration of the problem using a combination of built-in platform intelligence and intuitive visualization.

Additionally, Strobe integrates with the industry leading end-to-end APM tools AppDynamics and Dynatrace, both of which provide views of individual transactions as they flow from the end-user’s device to through the different layers of servers and back.

The AppDynamics integration enables users to capture SQL Analysis data on SQL that was initiated off platform to resolve issues with applications that use DB2 for z/OS. AppDynamics captures the JDBC call to DB2 but does not see what happens in DB2. Strobe for DB2 allows the user to analyze data with iStrobe’s SQL Analysis feature and perform “what-if” analysis for improving it. All this can be done in the absence of a Strobe measurement, which provides almost no-overhead monitoring for DB2 DDF.

The dynaTrace integration provides the ability to capture high-level data on the business transaction as it flows from CICS region to CICS region in z/OS. Analysts who see a problem can request a Strobe measurement from the Dynatrace UI to provide root cause analysis of why a CICS transaction may be using more CPU.

Reduce Time and Resources Wasted in War Rooms

Too much time and too many resources are being wasted in “war rooms” where it may take many hours or days to find the root cause of a critical performance problem. Strobe quickly and easily pinpoints the root cause of performance problems even in complex environments with multiple languages, databases or teleprocessing subsystems, putting an end to finger-pointing.

Enable the Transitioning Mainframe Workforce

As experienced IT staff transition into retirement and new inexperienced employees come on board, a large amount of knowledge will be lost. Strobe provides “the answer” up front, supplying the precise, in-depth application performance data needed to locate and eliminate sources of excessive resource demand. Built-in intelligence reduces the reliance on human expertise and experience.

Integration with Compuware’s modernized mainframe development environment, the Eclipse-based Topaz Workbench, means you can also initiate Strobe measurements for both active and future application instances from within a modern open development environment.

Analyze Which Mainframe Application Performance Problems to Fix

Strobe presents a detailed analysis of the most important tuning opportunities found. It quickly identifies performance problems that are the simplest to fix as well as the top five opportunities in major categories such as COBOL, DB2, Language Environment and file access.

Using the built-in analytics of iStrobe Strobe Insight, users can execute the reports to look for problems across many profiles or query SMF data to determine what jobs should be measured.

Measuring and Reporting on Mainframe Application Performance Management (APM) Success

Strobe Insight provides analysis for all major subsystem components, including batch, CICS, DB2 and IMS. Performance data can be stored in a Microsoft SQL Server, Oracle or DB2 database. Performance data can be stored in Microsoft SQL Server, Oracle or DB2 databases. Using the Business Intelligence Reporting Tool (BIRT), you can either view the supplied reports or easily craft new custom reports to view your data from a variety of perspectives. This allows you to follow trends, identify the best areas for tuning efforts, and isolate response time and CPU consumption problems.

Deep-dive Java Performance Analysis

Strobe seamlessly integrates with Topaz for Java Performance giving staff the ability to take comprehensive measurements of their whole Java environment. Strobe delivers valuable insight about Java Virtual Machine (JVM) jobs and specialty processor usage, while Topaz for Java Performance gives an inside view of the JVM for targeted tuning of Java programs.

What’s New in Strobe?

Strobe for Deep Mainframe Application Performance Analysis