October 7, 2016 | Written by: Ben Stern
Share this post:
The development and implementation of business applications has dramatically changed in recent years. Companies are moving to the public/private hybrid cloud, using docker containers and adopting platform as a services and moving to DevOps. All of these changes have driven significant changes in the tools used to monitor and manage applications. As applications have become more dynamic, traditional monitoring tools do not provide enough visibility, driving the need for more effective application performance management (APM).
The mainframe remains
One thing that hasn’t changed over the years: companies continue to rely on the mainframe. The mainframe continues to run the most mission-critical, time-sensitive, and highly-available business applications. In some cases, the mainframe applications haven’t changed, but business applications have evolved. Companies are using modern application paradigms for the front end of applications and tying those into existing backend infrastructures via APIs. The back end infrastructure consists of mainframe components such as IMS, CICS and DB2.
Monitoring in silos is ineffective
Traditionally, companies have taken a siloed approach when it comes to mainframe and distributed monitoring. The owners of distributed systems assumed the mainframe team would ensure the availability and performance of the back end systems. Companies have used IBM OMEGAMON monitoring capabilities to provide visibility into back end systems and other tools to monitor the distributed platforms.
That approach worked in many cases because business applications were very static. In today’s world, applications are more dynamic, which has driven the need for better visibility. For example, as users attempt to debug an application response time problem, they can’t ignore the back end infrastructure running on the mainframe. Why? The infrastructure may be the root cause of their performance and availability issues.
The monitoring solution for dynamic applications
IBM has introduced tools that can provide companies with complete visibility into today’s dynamic business applications. IBM OMEGAMON for APM allows companies to use their existing OMEGAMON monitoring agents and integrate them with IBM APM.
By providing integration with existing OMEGAMON infrastructure, no additional overhead needs to be added to the mainframe environment. It also provides organizations with a truly end-to-end view of their business application. Users can log into a web-based user interface and access the advantages of the APM capabilities.
How it works
IBM APM uses the following techniques to detect, diagnose and avoid problems:
- Web response time monitoring. If response time is good, issues on the back-end are a secondary concern. By monitoring web traffic, IBM APM provides response times for individual users, specific geographic location and more.
- Synthetic transaction monitoring from global locations is your first line of defense. You can utilize 15 points-of-presence (PoP) servers across the globe and supplement the data with your own PoP servers to monitor potential performance from different locations. Synthetic transactions also monitor whether service-level agreements are being met.
Synthetic transactions run from multiple locations
Pinpoint a bottleneck by monitoring transactions as they flow through the application.
Once the bottleneck is identified, users can drill down into the details. In the case of mainframe components, users can drill down into the mainframe and distributed components.
Top-level view of CICS health and performance.
Top-level view of z/OS health and performance.
IBM APM applies sophisticated analytics to the logs and metrics. For detailed information, see IBM Operations Analytics Log Analysis and IBM Operations Analytics Predictive Insights.
A predictive insights anomaly.
Can your company afford its distribution and mainframe teams to be working in silos? Learn about IBM OMEGAMON for Application Performance Management here.