CONICIT USER INTRTFACE
ConicIT has few components which expose UI for different purposes
CONICIT WEB UI – UNIFIED PRESENTATION OF DATA FROM ALL MONITORS
ConicIT Web UI is the main UI for delivering the unified presentation of performance information, including tables, graphs, and comparative graphs. In ConicIT Web UI you can see a list with all of the investigations that ConicIT has done during the last year. Each type of investigation is typically performed once a minute, so you get a high resolution of information.
You can choose to see the details of the most recent investigations or of investigations from any time in the last year. In each investigation you will see tables which clearly show some perspective of performance data, such as CPU, transaction-rates, memory consumption, queues, response times, wait-percents for resources, and many more. In addition you can click on each one of the metrics and get the graph of that metric, including a green zone showing the expected-values for that specific metric during that time. When watching graphs you can choose any time-span you wish, and you can choose to see few figures on the same graph, or to compare the graphs of the same variable during different days.
With a single click you can get to see each investigation, and each investigation can show aggregation of z/OS performance data from few monitors, such as few TMON-CICS, combination of different type of metrics, or aggregation of information per group (e.g. sum of transaction-rates per group of few CICSes. In addition ConicIT can show calculated variables such as ration, the change (delta) of a variable from previous minute, or any other calculated metrics.
ConicIT customers testify that thanks to ConicIT unified presentation, they stopped holding dozen of screens constantly showing information from various mainframe-monitors. With ConicIT one screen is enough, and ConicIT is doing for you the hard work of following the actual data. When something goes wrong ConicIT will make sure that you’re the first to know, much before you get calls from customers care department.
CONICIT BRINGS SPECIFIC MAINFRAME PERFORMANCE DATA TO SPLUNK EFFICIENTLY
EASILY BROWSE THROUGH MAINFRAME PERFORMANCE DATA
On the left you can choose the type of data you want to see along with the exact minute that you’re interested in.
In this example we chose to see real-time CPU consumption of started tasks. These metrics are actually calculated values that ConicIT produce based on total CPU-Time of each started task from IPL, because monitors like TMON and Omegamon doesn’t provide the real-time CPU consumption.
SEE GRAPHS OF MAINFRAME PERFORMANCE METRICS
In this example ConicIT is showing a graph of transaction-rates of a CICS during eight-hours. You can see the graph along with the green range which shows the expected range for that variable at each hour based on the behavior on the previous weeks. In case of a strong enough deviation from the green range and based on machine learning algorithms, ConicIT will send alerts to notify you about the suspicious incident.
ConicWatch is a Windows application that runs in your system tray and shows you the alerts from ConicIT in real-time. Whenever an alert is sent from ConicIT, ConicWatch icon is turning red and can present a pop-up notifying about the new alert that have arrived. Every IT personnel can install ConicWatch and can choose the list of topics and systems that he’s interested in. So if some user is only responsible for CICS for example, he can choose to get only the CICS alerts while another user might prefer to see only the DB2 related alerts.
When you open ConicWatch you can see a list of alerts that were received recently. If you click on an alert, you’ll be forwarded to the relevant page in ConicIT Web UI showing details about that alert, and you can also ask to see the relevant graph.
GET ALERTS ABOUT MAINFRAME PERFORMANCE TO YOUR PC
Whenever ConicIT recognizes a problem with the mainframe performance, it sends the alert to ConicWatch – a small Windows application on your PC. You can configure ConicWatch to popup the alerts, so you’ll always know when a new alert arrives.
ConicAdmin is a Windows application that allows you to control the behavior of ConicCollect software, and to extend its capabilities. ConicAdmin is not just a configuration tool, but it allows you to add new analysis to ConicIT, with special wizards. You can define how to connect to the system (e.g. the monitor of your mainframe), how to perform the specific command (e.g. receive CPU of CICS), and define how to parse the results using our special parsing-wizard. Once you finish the parsing wizard, you define how ConicIT should analyze the data and when it should send alerts. You don’t need to work hard here, because recently ConicIT introduced our new Auto Rule Generator wizard, which guides you with detailed questions about how you wish to treat each variable, and for which variables you want alerts.
ConicIT will automatically apply its machine learning algorithms to automatically recognize z/OS performance anomalies on any variable that you’ve asked to have alerts about. The flexibility of ConicIT to create new commands starts from its architecture which embeds Perl-code for applying the analysis. For each command, ConicAdmin defines the Perl-code that implements the business logic. The Perl-code is automatically generated at the end of the Automatic-Rule-Generator Wizard, but then you can watch the Perl-Code and even change it if you wish.
PARSE THE MAINFRAME MONITOR SCREEN INTO PERFORMANCE VARIABLES
In this example ConicAdmin let the user define the parsing of TMON screen with performance metrics of CICSes. Using the mouse, the user adds variables and mark the columns where they appear. The result is that ConicIT will turn each such screen into variables of CPU, transaction-rates, and transactions-average response time per CICS.
THE AUTOMATIC RULE GENERATOR WIZARD
During this wizard, ConicAdmin asks you series of questions like which is the ID field for each line, and which are the numerical variables. It will allow you to define variables for which you want to have a Delta-variable (the way it change each minute compared to the previous one, important for example to get
the real-time-CPU based on cumulative CPUTime). It allows you to set ratio-variables (the ratio between two variables), and allows you to define for which variables you’ll see graphs, have static alerts, and have proactive alerts based on dynamic thresholds and machine learning.