SAP Analytics Cloud(SAC): One platform for all business needs

SAP Analytics Cloud is a product where companies can bring together?analytics and planning?with unique integration to SAP applications and?smooth access to heterogeneous?data sources to meet their business needs. SAC can also be used to predict customer behavior and stay ahead of the competition with predictive analytics.

What can we do with the SAP Analytics Cloud?

Performance Challenges

SAP Analytics Cloud enables to perform sophisticated analysis on large volumes of intricate data. These complex scenarios can sometimes lead to less-than-ideal performance times for end-users.?

• Experience performance Issues.
• Refresh quality.
• Slowness while dashboard execution.
• Access to SAC application itself is slow.
• Limitation of data records being shown.

Below 3 components are impacted by several different reasons:

1. Client Time (frontend) | 2. Network Time | 3. SAP Analytics Cloud service time (backend)?

Solution Approach

By using SAC inbuilt performance analysis tool, one can identify the problematic parts in business scenario and see how much time is spent in which component (frontend, network, backend).

Note:

1. This tool only traces the performance of stories, data actions and apps (applications written with SAC's application designer).
2. Other activities and navigation actions such as opening the SAC modeler, etc. are not traced.

1. Performance Benchmark Tools

• While client and network performance are addressed in the?SAP Analytics Cloud Performance Benchmark, SAC is considered by some people as a Blackbox?regarding the widget’s performance. SAP acknowledged the need to provide additional information in SAC, to help the designers to understand the widget affects on dashboard performance and how they can have the most efficient implementation.?
• The Performance Benchmark helps to measure client hardware score and network speed. These have a measurable influence on SAP Analytics Cloud user experience.

2. Performance Statistics and Analysis

The Story helps not only to address Backend, but it also lightens the dark of some SAC Frontend related Information and this again is a sample story that can be adjusted as per the designers needs.

3. Performance Analysis Tool?

• The performance analysis tool is an analytical application which allows easy analysis of the metrics provided by SAC_PERFORMANCE_E2E?
• The performance analysis tool helps to analyze a single story or analytic application run?through different search criteria

In the overview section one can see – the user details, page details available in the story, Top5 widgets across all the pages, models backend runtime and lastly the comparison between Median and Maximum Runtime distribution per processing layer.

Processing information

Processing Information will provide the views with respect to different processing:

1. Page Load Time?

2. Widget Drilldown?

3. Runtime Distribution

4. Time Series Charts

Data action Performance Statistics and Analysis

The goal is to enable the user to:

• Get an overview of the Data Actions.
• See statistics, detect patterns and trends in the execution history.
• Analyze single Data Actions that have been triggered.

Best practices for building efficient SAC dashboard

• Refresh Browser Cache.
• Check if "Optimize Story Building Performance" option is enabled in the Model Preferences under Data and Performance.
• Whenever possible, choose to?show unbooked data in a chart. This means that SAC?has to spend less time differentiating between booked and unbooked data.
• Avoid specifying Exception Aggregations in the Model and instead, use the?Restricted measures?or?Calculation?functionality in stories.
• Try designing the story using?Responsive pages?instead of Canvas or Grid pages. Responsive pages allows story content to re-flow depending on the size of the screen it is being viewed on.
• Use pages to break up the story by category or type of information. Put most-viewed content on the first page to make it easily accessible.
• To ensure performance while working with blended data,?avoid creating?Linked Dimensions?on?Calculated Dimensions.
• Use relevant filters to limit the number of records in the story to one million or less.
• While creating a story with many elements based on the same information, try?adding?story filter?capabilities instead of individual filters for pages, charts, or tables.