Ensure z/OS Performance with Predictive Intelligence
The goal of IntelliMagic Vision is to identify potential issues before they impact applications. This is achieved by an intelligent rating system where workload measurements are combined and compared with the capabilities of the systems. The resulting ratings are shown in dashboards that represent the health of your z/OS infrastructure.
There are separate dashboards for areas, such as:
The dashboard indicates infrastructure health by showing the key ratings for that part of the environment. Dashboard ratings are based on the analysis of thousands of underlying data points, making the dashboards a very dense summary of hundreds of charts. The result is that issues and risks are flagged proactively before application performance degrades.
Watch the video below for an in-depth look at our Dashboards and Ratings.
Benefits and Capabilities
Overall z/OS Performance & Health Status
The dashboards show the summary; drill-downs allow you to explore details. The highest level dashboards are color coded bubble dashboards, as shown below. The color and size of these bubbles show the ratings for the underlying metrics: a green bubble indicates a healthy situation, a yellow bubble indicates that a problem is developing, and the large red bubbles flag more severe risks or issues. With these visual cues, it is extremely easy to see the health status of the entire environment at a glance. IntelliMagic Vision can be set up to send emails automatically when a dashboard contains one or more yellow or red warnings.
Tip! Read this white paper: z/OS Storage Dashboards with IntelliMagic Vision
Investigate Performance Details
The bubble dashboard shown above provides the most compact view, but the rating is based on a very detailed level of analysis. To get more information about an issue, you can click on the bubble to get to to the next level of detail. The bubbles are replaced by mini charts that show the metrics over time, as well as the rating and threshold values.
Root Cause Analysis
When the dashboard shows a problem, you can click on one of the mini charts to get a full version of the chart, which also contains an explanation of the metric and thresholds, as well as recommendations on what could be done to address the issue. The border of the charts is colored in the same way as the bubble dashboard: a green border means a healthy metric, yellow is for early warning and red indicates a larger issue.
Each individual chart contains multiple drill-down options to go to the deepest level of detail in any direction, for example to find the individual RAID array that was so busy that it caused a large red circle for drive response time in the highest level dashboard.
There are thousands of pre-defined charts and reports available in IntelliMagic Vision, grouped into logical sets. If you are interested in showing a combination of metrics or filters that is not available out of the box, you can customize the charts, or define your own from scratch and add it to your favorite chart set. The thresholds that are used in the rating system are also customizable to fit your situation.
All charts and reports can be exported to CSV, HTML, PDF, PowerPoint and Splunk.
Many metrics are best shown as line or area charts, but some values are better looked at in a different fashion. The variance chart, for instance, is a great way to show (im)balance. In the example below, you see the FICON throughput per port. A green dot indicates average throughput for the port, a green rectangle that shows the standard deviation and a yellow area that shows the minimum and maximum value over the entire period.
Is Your Car or Mainframe Better at Warning You?
Early warning can help avoid problems before they occur or impact production, and help fix a minor issue before it gets worse.
AIOps, Mainframe Monitoring, MLC Optimization, Oh My: Hot Mainframe Topics of 2018
As 2018 nears its close, I wanted to reflect on the world of RMF/SMF performance analysis and provide an ode to the topics that resonated most within the industry.
Achieving CPU (& MLC) Savings on z13 and z14 Processors by Optimizing Processor Cache
Beginning with z13 processors and continuing with z14s, customer experiences have confirmed that delivered capacity is more dependent than ever before on effective utilization of processor cache.