Monitoring Resources in SAP HANA Cockpit

Objectives
After completing this lesson, you will be able to:

After completing this lesson, you will be able to:

  • Use the SAP HANA Cockpit resource monitoring cards

SAP HANA Cockpit - Services Card

Lesson Overview

In this lesson, you will learn how to monitor the SAP HANA Cloud database services using the SAP BTP version of the SAP HANA cockpit.

You will also learn about the different features available in the Performance Monitor app.

In addition, you can identify which tables have a high memory usage and have a high read/write load.

Business Case

As an database administrator, you want to monitor the availability and resource consumption of the SAP HANA Cloud database services.

SAP HANA Cockpit - Services Card

See the following video to find the SAP HANA cockpit - services card.

The Manage Services app gives you an (1) overall status summary of your database.

The Reset Memory Statistics button (2) lets you clear the statistics history.

Peak used memory is the highest recorded value for used memory since the last time the memory statistics were reset. This value is useful for understanding the behavior of used memory over time and under peak loads. Resetting peak used memory allows you, for example, to establish the impact of a certain workload on memory usage. If you reset peak used memory and run the workload, then you can then examine the new peak used memory value.

The Go to Alerts button (3) lets you directly open the Alerts application to investigate the alerts associated to this resource.

The Kill Service button (4) stops the selected service immediately and if the related option is selected, creates a crash dump file. Normally the SAP HANA daemon will restart the killed service.

The Settings icon (5) lets you add and remove columns from the display. Not all columns are visible by default.

The Service column (1) show the service name, for example, indexserver, nameserver, xsengine, and so on

The Status column (2) show the The status of the service. The following statuses are possible:

  • Running - the service is operating within normal parameters.
  • Running with Issues - at least one service is not running, or there is at least one high alert.
  • Starting - the service or system is starting.
  • Stopping - the service or system is stopping.
  • Stopped - the service is stopped.
  • Not Running - to investigate why the service is not running, you can navigate to the crash dump file created when the service stopped.

The Role column (2) shows the role of the service in a failover situation. Currently not applicable in SAP HANA Cloud.

The Port column (3) show the port number that the system uses for internal communication between services

The Process ID column (4) show the operating system process id assigned to this service.

The CPU column (5) displays a mini chart visualizing the CPU usage of the service. Choosing the mini chart opens the Performance Monitor app for a more detailed breakdown of CPU usage.

The Memory column (6) provides a mini chart visualizing the memory usage of the service. The different colors represent the following:

  • Dark green shows the service's used memory.

  • Light green shows the service's peak memory.

  • The grey stroke represents the effective allocation limit.

  • The light grey background represents the physical memory.

Choosing the mini chart opens the Memory Analysis app for a more detailed breakdown of memory usage.

The SQL Port column (7) show the SQL port number assigned to this service. Not all services require an SQL port.

Selecting the mini chart in the CPU Column (1), the Performance Monitor app is opened and displays a detailed breakdown of CPU usage over time.

Clicking the Memory column (2) mini chart opens the Memory Analysis app for a more detailed breakdown of memory usage.

The applications Performance Monitor and Memory Analysis will be explained in detail in their respective SAP Learning Card.

Use the Services Card

SAP HANA Cockpit - Performance Monitor Card

Business Case

As an database administrator, you want to monitor the database performance by checking the memory, CPU and disk resource consumption of the SAP HANA Cloud database.

SAP HANA Cockpit - Performance Monitor Card

Analyzing the performance of the SAP HANA database over time can help you pinpoint bottlenecks, identify patterns, and forecast requirements. Use the Performance Monitor to visually analyze historical performance data across a range of key performance indicators related to memory, disk, and CPU usage.

In the SAP HANA cockpit - Database Overview page, search for the Monitoring Performance app (1).

The Monitoring card (2) shows several links to useful monitoring applications like the Monitor Performance and the Monitor Table Usage app.

Clicking on the Monitor Performance link opens the Performance Monitor app which, by default, gives a detailed breakdown (3) of several important resources (CPU, memory, transactions and threads) of the SAP HANA database over a time frame of the last hour.

Monitoring the performance of your SAP HANA database is important. Performance can be measured on the basis of memory, CPU and disk. All these KPIs are important and could be used to investigate possible performance issues.

To make the data as insightful as possible, the SAP HANA cockpit provides three cards that display a mini graph on their specific KPI (memory, CPU, and disk). All three cards open the Performance Monitor, but will immediately focus on their respective KPI.

Hint
All the features shown in the following screen shots are part of the Performance Monitor app and are independent from the selected resource memory, CPU or disk.

The Performance Monitor opens displaying the load graph for the selected resource: CPU, disk, or memory. The load graph initially visualizes resource usage of all hosts and services listed on the left according to the default KPI group of the selected resource.

By default the Performance Monitor app show the measurement values of the selected resource from the last hour.

Defining the monitored time frame (1) by entering your desired dates or selecting form Presets (2). The following presets are available:

  • Last 5, 15, or 30 Minutes
  • Last 1, 3, 6, 12 or 24 hours
  • Last 2 or 7 days
  • Today, this day last week, yesterday, day before yesterday
  • This week, previous week or this month
Note
Using the auto refresh button, you can set the auto refresh rate between every 10 seconds to every 10 minutes.

Using the My View (1) drop down menu to quickly switch between the different resources. The four default views (CPU, Memory, Disk and Default) can be used, but you can also create and save your own views with the monitoring KPIs of your choice.

Using the manage configurations (2) button allows you to select your own KPIs.

In the Add and Remove KPIs pop-up window, you can select from which services you want to see the selected KPIs measurement data. You can also select several on host level aggregated KPIs values.

Create a complete snapshot of the current performance monitor data by exporting the monitoring views data into the a LOAD_HISTORY.zip file. The content of the following system views will be exported:

  • SYS.M_LOAD_HISTORY_INFO
  • SYS.M_LOAD_HISTORY_HOST
  • SYS.M_LOAD_HISTORY_SERVICE
  • SYS.M_DATABASE
  • SYS.M_SERVICES

Use the Performance Monitor Card

SAP HANA Cockpit - Monitor Table Usage app

Business Case

As a database administrator, you want to know which tables have a high memory usage and have a high read/write load. With this information you can identify if tables could be moved to the SAP HANA Cloud, Data Lake.

SAP HANA Cockpit - Monitor Table Usage app

In the SAP HANA cockpit - Database Overview page, search for the Monitoring (1).

The Monitoring card (2) shows several links to useful monitoring applications like the Monitor Table Usage app.

Clicking on the Monitor Table Usage link opens the Table Usage app which, by default, gives a detailed bubble chart Top 50 (3) of the table usage by read and write access.

Monitor table activity to identify where you can improve performance and reduce memory utilization. Large in-memory tables that are accessed infrequently are good candidates for the SAP HANA Cloud, Data Lake option.

Use the Go to Alerts (1) button to see if exceptional large table with very high read/write access cause alerts in your SAP HANA Cloud database.

Filter Tables (2) allows you to filter by the number of times a table is accessed and the table size. You can also specify how many tables you want to see in the listing.

The bubble-chart and table-chart buttons (3) let you switch between the graphical view and the list view.

In the Filter Tables (1) select options field, you can combine all the filter options. This means you can filter the Top 50 table list to only shown tables with a small number of read/write accesses, but are big in their size.

The example in the above screen shot shows the Top 50 list of tables that have less that 5000 reads and are bigger then 150 MB. The list shows 4 tables. These 4 tables might be candidates for moving to the SAP HANA Cloud, Data Lake, but that needs to be discussed with the application and the development team because the way the tables can be accessed will change when moved to a data lake.

Use the Monitor Table Usage Card

Save progress to your learning plan by logging in or creating an account