Enterprise observability and cross-platform cross-application performance anomaly diagnosis
Background
In today's complex environment, enterprises often have hundreds or thousands of applications running simultaneously and interactively to achieve business goals. At the same time, all business units expect better application performance and end user experience.
With the increasingly complex modern application environment, cross-platform applications also bring great challenges to operation management. However, existing technology tools struggle to provide consistent and efficient system visibility, requiring multiple teams to investigate various parts of their infrastructure and applications to determine the root cause when a problem exists. IT operations teams and application development teams cannot obtain a single source of truth, often encounter inconsistent alarms, late or false positive events, and spend a lot of time and resources on problem location.
In a rapidly changing and complex IT environment, automatic, real-time and continuous application performance observability is more and more becoming a basic requirement for a new generation of monitoring and application performance management solutions.
IBM Instana APM
As a leading fully automated application performance management solution, IBM Instana provides six unique capabilities: automatic discovery and monitoring, contextual guidance, application perspective, Pipeline feedback, root cause analysis, and borderless analysis, helping enterprises deal with management to meet the challenges of managing complex application environments in the hybrid cloud architecture, to achieve integrated monitoring of infrastructure and applications.
The six unique core capabilities of IBM Instana are:
• Automated discovery and monitoring. Instana provides automatic application performance management capabilities such as automatic discovery of application architectures and automatic analysis of call chains. It can automatically, real-time, and continuously discover complete servers and application components, automatically trace all requests, and accurately understand application performance status and problems in a timely manner.
• Automatic correlation of contextual information. Instana automatically correlates calls between applications, as well as applications and their associated underlying infrastructure, such as middleware and database server dependencies, thereby providing real-time, interrelated and interactive dynamic models. Instana can also intelligently monitor application performance metrics anomalies and analyze their impact scope, helping application maintenance personnel to quickly locate and solve problems.
• Application perspectives. Instana provides an intuitive and unified observation dashboards, and can customize the application view, to centrally display all components, services and interaction requests called by the application, and realize integrated display, analysis and alarm.
• Unbounded analytics. Instana analyzes all user requests from different angles to quickly find and solve performance bottlenecks, provides unlimited flexibility to trace all links and performance data from the entire application request, intelligently analyzes every event that affects services, and provides end-to-end track data to ensure the health of the application.
• Root cause analysis. Instana provides application full-link tracing capabilities, and automatically performs event correlation, anomaly detection, alarm, and root cause location based on dependencies and expert experience. The Instana machine learning model can also actively mark the root cause of the problem and give suggestions for problem analysis in advance.
• Pipeline feedback. Instana associates application version releases with infrastructure and service performance data, automatically analyzes build and deployment pipeline events, and discovers, locates, and resolves problems caused by new application version releases in a timely manner.
In June 2022, Instana has been named as a LEADER in the new 2022 Gartner Magic Quadrant for Application Performance Management(APM) and Observability. This achievement relies on extending APM Enterprise Observability to IBM mainframe systems, in addition to strengths such as making analytics and AI-ML key elements of the solution, building specialized tools for "modern architectures" (i.e., cloud-native), and simple pricing model.
In addition, IBM Instana has won the Devies 2023 Award in the Developer Category for Observability and Monitoring. It commends Instana for building innovative solutions to meet the complex demands of our mainframe users.
Instana supports over 250 technology stacks, enabling automatic discovery and monitoring of different technology stacks in hybrid cloud environments, providing the most comprehensive enterprise observability. IBM mainframe platforms and applications remain crucial component in today's hybrid cloud environments.
Instana extends application performance monitoring to include transaction tracing for IBM mainframe applications, directly integrating performance metrics data from IBM Z OMEGAMON and SMF. This achieves end-to-end enterprise observability from distributed microservices applications to mainframe applications, middleware, databases, and the z/OS operating system.
Even application owners without experience in mainframe platform operations can quickly view key performance metrics and system dependencies of mainframe services called by their applications within the context of their application. They can analyze the root cause of application performance issues through application call hierarchies.
Once the root cause analysis narrows down to a specific application transaction and mainframe subsystem, application owners can leverage the transaction or system data automatically discovered by Instana to connect with the corresponding mainframe domain experts. This allows for rapid identification and isolation of potential issues on the mainframe, shortening the time to resolve problems and minimizing the impact on business application performance and availability.
Use Case introduction
Now let’s take a look at the IBM zAIOps solutions and use case: Enterprise observability and cross-platform cross-application performance anomaly diagnosis
With the increasing complexity of application architectures in enterprise environments, cross-platform application components are commonly interconnected. However, traditional application monitoring tools lack full end-to-end observability capabilities for cross-platform applications. Additionally, for some organizations, the IT resources are shared across applications. For example, the mainframe database resources are usually accessed by CICS applications, batch applications, and DRDA applications simultaneously. An error in one application may cause performance issues in other applications. It is challenging to detect and diagnose such problems using traditional performance monitoring and analysis tools.
By leveraging the IBM Instana on z/OS solution, you can achieve end-to-end observability and performance monitoring for cross-platform application service calls. It also helps IT operations quickly identify performance impacts caused by shared resources conflict between applications and enables rapid problem analysis and diagnosis.
This use case demonstrates the use of IBM Instana and Instana on z/OS to monitor cross-platform business application systems, including:
- Transactions entering CICS® Transaction Server for z/OS and Db2 through HTTP or SOAP distributed entry points.
- Java applications that use JDBC to access Db2
These two types of applications share the same table (CBANG table in Db11) in Db2 on z/OS, and there are performance impacts between them.
The software components used in this use case demonstration include:
- IBM Instana on z/OS: Extends Instana's observability capabilities to the mainframe, including CICS, Db2 on z/OS, DRDA, MQ, CTG on z/OS, IMS, z/OS Connect, enabling end-to-end enterprise observability.
- OMEGAMON for z/OS: Enables real-time monitoring of critical metrics for the z/OS system.
- OMEGAMON for CICS: Enables real-time monitoring of critical metrics for the mainframe CICS subsystem and transaction performance.
- OMEGAMON for Db2: Enables real-time monitoring of critical metrics for the mainframe DB2 subsystem and transaction performance.
- JDBC for DB2 on z/OS: Enables distributed Java applications to access the mainframe DB2 database using the DRDA protocol.
Lastly, this use case also demonstrates Instana on z/OS's ability to trace several important transactions on the mainframe, including MQ, CICS Transaction Gateway, z/OS Connect Enterprise Edition, and more.
Demonstration
Now, let’s see a demonstration of this use case.
Summary
At last, let us sum up the above use case.
This use case demonstrates the use of IBM Instana on z/OS to trace all transaction chain data using automatic discovery technology. It automatically constructs a complete transaction call chain topology from distributed microservices applications to mainframe CICS transaction server and Db2 databases. It intelligently detects transaction performance anomalies and associated metric abnormalities through smart alerts, promptly identifying performance impacts between applications. It enables quick identification of nodes and root causes of issues using transaction trace analysis and performance metric data.
It provides an end-to-end enterprise observability solution from distributed microservices applications to mainframe applications, middleware, and databases. IBM Instana on z/OS enables comprehensive enterprise observability. It monitors applications and system resources in hybrid cloud environments through a single interface, providing high-fidelity data, real-time automated infrastructure monitoring with 1-second metric granularity, real-time metric collection, and prevention of application bottlenecks and performance issues, among other key capabilities.
It helps operations and development teams understand the interplay of application services in hybrid cloud environments, enabling quick issue identification and resolution, and improving operational efficiency. Through built-in intelligent actions, Instana assists application owners in quickly resolving application bottlenecks and optimizing performance. Application maintenance and mainframe operations teams no longer need to expend significant time and resources in "war room" to troubleshoot root causes. It eliminates monitoring blind spots caused by disparate monitoring dashboards, and a lack of correlation in monitoring data, ensuring the secure and stable IT operation.
Watch the complete video for this use case.