April 15, 2018

Platform

New Features:
  1. Improved repository management for multi-tenancy.
  2. De-couple the edge API with inner API for better management and monitoring.
  3. Introduced service routers to effectively handle communication with inner API and edge API layers.
  4. Each plugin service has separate JVM runtimes to enable incessant deployment and monitoring.
  5. Load balance for core service layers.
  6. Enhanced the event service layer and consumer layer to improve agility and scalability.
  7. Memory optimization and performance improvement for repository persistence layer.
  8. SFTP Connector for CSV and Excel files.
Enhancements:
  1. NLP functionality has been revamped.
  2. The enhanced user interface for Data Center, Home, and User Management modules.
  3. Administration Module- The following enhancements are given for the ‘Predictive Settings’:
    1. Multiple R Server Connections- Users have provision to choose separate R servers for scheduling and workflow processing
    2. Multiple Spark Server/cluster connections
    3. Multiple Python Server connections
    4. Provision to integrate with R Bokeh charts
    5. Enable/Disable parallel processing in R along with a number of cores
Notes: API Data Connectors, Natural Language Search, and Play Plugin are currently in Beta.

Business Story

New Features:
  1. Business Story has been provided as a plugin in the platform app menu.
  2. Formula Field Editor- Feature to create a calculated field based on fundamental mathematical formulas
  3. Range- Feature to create range-based bin as Dimensions, from Measures
  4. Responsive legends are introduced.
  5. Manual Sort (order by) is now possible for dimension.
  6. ‘Quarter’ and ‘Week’ are provided as date intervals in the ‘Series Properties’.
  7. Datastore performance is improved.
  8. A business story can be refreshed based on an external event.
Enhancements:
  1. PDF Export has been improved
  2. NLQ Enhancements:
    1. Auto-suggest feature to display correctly structured queries
    2. Notification for the unidentified keywords
    3. Enhanced search options- Month value has been included
    4. Improved appearance of the suggested chart
    5. Data Mining is introduced in Elastic Search
    Note: NLQ supports all types of Date Formats

Data Preparation

New Features:
  1. Filter component with ‘ANY’ condition has been added.
  2. Inferring date & date time formats are enabled in the Datatype component.
  3. Cassandra and HDFS writers are introduced as output components.

Data Pipeline

New Features:
  1. Big Data Pipeline framework is developed into a full-blown reactive Big Data Pipeline.
  2. It is an entirely automated system with orchestration support by Kubernetes, data channel support by Kafka, and data distributed processing support by apache spark.
  3. Big Data Pipeline can take data from different sources like web socket, databases, Datamart & analytical stores.
  4. It can deliver data to different endpoints like Dashboards, Business Story, website, and data stores including distributed data stores like HDFS, ES.
  5. Big Data Pipeline supports real-time, Event-based (near real-time) and batch mode data.
  6. It supports cloud and on-premises infrastructure.
  7. Ability to run custom Pig scripts, Hive scripts, Store procedures, Talend job, Unix Shell scripts, BDB Data Preparation jobs and Predictive services.
Note: The Data Pipeline Plugin is currently in alpha.

Dashboard Designer

Charting
New Features:
  1. KPI Tile is added as a new component
  2. Custom Tooltip is available for Area, Line, Bar, Bubble, Column, Mixed, Scatter Plot, Histogram, Time-Series and Group Bar charts.
  3. Data labels on charts: Number formatters and alignment options are provided
  4. Axis Markers:
    1. No duplicate y-axis markers when Auto unit and default precision are pre-defined
    2. Number of Y-Axis markers are auto-calculated based on the height of the charts
    3. X-axis labels can be skipped using the script
    4. Currency formatters like $ should be placed after (-) negative sign
    5. Currency formatter can display IND RUPEES Symbol and configure values into Lac/ Cr
  5. Action Palette option has been provided for the Bar, Time Series, and Pie charts.
  6. Export as Excel: Numeric values can be exported as a number in excel with the details of the applied filter conditions.
  7. Color from drill functionality now supports multiple charts
  8. Component property and script can control title box height and side paddings.
  9. Different pointer type option can be selected for clickable data points.
  10. Tooltip precision, percentage, and cumulative sum display can be controlled for Column, Bar, Group Bar, Mixed, Time-Series and Area charts.
  11. Multiple data points can be displayed at the same position, but tooltip will highlight the topmost data point
  12. Number Formatter property is added in the component formatter properties to select a thousand separators in the axis markers.
  13. Show data: Currency and number formatters are enabled
  14. Component maximize and minimize icons can be changed from script
  15. Data Palette Properties Enhancements:
    1. Scattered Plot Chart:
    2. The ‘Radius Field,’ ‘Color field,’ ‘Other field’ (to select ‘y’ or ‘x’ axis) are provided via a pop-up window
    3. Other-Field display name and Color-Field display name are grouped into advanced properties section which can be configured
    4. Solid and gradient filling option.
  16. Bubble Chart:
    1. The ‘Fixed Radius field,’ ‘Color-Field,’ ‘Radius-Field’ are provided via a pop-up window
    2. Solid and gradient filling option.
  17. Time-Series Chart:
    1. The default position of the slider can be set to right or left side.
    2. Chart animation has been provided to highlight data point for column/line types
  18. Group Bar & Bar Charts: Stacks drawing is displayed top to bottom
  19. Treemap: Number formatted data is supported in data display and data labels
  20. Legend:
    1. Initial visibility status of series fields can be checked from legend component
    2. Time-Series: Fluid Legend is added to display the legend in top/bottom/left and right
  21. Combo Filter: Title properties are added
  22. Radio Button: Users can change the view type as Button-Set or Radio-Set view via the properties
  23. Data Grid:
    1. Multiple row selection in the drill is allowed
    2. More shapes are added in comparison indicator
    3. Fit Column property is added to set all fields in offered width/ Users can define fields width and make horizontally scrollable
    4. The indicator can support comma separated data
    5. Sorting of fields can be controlled through Dataset properties
    6. Tooltip can be enabled for fields without applying Alerts
  24. Data Service:
    1. Empty filter keys can be passed in SOAP envelope and encode the inputs before calling the ‘getRecords’ service
    2. Dashboard performance: Endpoint listing service 'GetEndPointURL' has been implemented
  25. SDK Methods:
  26. The following SDK methods are added:
    1. ‘getCurrencyConversionRates’ -to get currency conversion rates in context
    2. ‘OnAfterRender’ callback method
    3. Methods to re-set field properties of components at dashboard runtime
    4. Methods to get widget object by name/ id/ type/ class for test-automation
    5. Methods to import Google fonts into the dashboard
Enhancements:
  1. SDK Methods:
    1. User Custom fields can be accessible in dashboard script
    2. Order of dataset operations can be configured

Designer

New Features:
  1. Option to design Mobile, Tablet, and Desktop view of a dashboard
  2. Data Store connector is added, and it can be consumed in Time-series, Bar, and Simple Data Grid components
  3. UI updates
    1. ‘Snap to Grid Size’ option is provided under Preferences
    2. Color picker UI has been improved to display ten last used colors and have theme colors
    3. Scaling type for the desktop dashboards can be selected via properties
    4. Version Number and Build Date are displayed on the Dashboard Designer homepage
    5. Data connectors and services list in ascending order
    6. The Component shadow properties can be controlled from the Preference
    7. The script window displays component and connection id
    8. Calculated Field: Rename and Delete options are provided
    9. Users can pin property palette on the right-side of the design area which can be hidden by clicking the ‘Close’ icon.
  4. Undo and Redo functions can be enabled for all the actions done on a component.
  5. Designer Group Management: Usability has been updated in the UI window.
  6. Users can select and move multiple components by dragging the mouse
  7. PA connector:
    1. Python Workflows are integrated into the Dashboard Designer plugin as a predictive service
    2. PA Web Service information (model ID) will be available in dashboard dependency details for migration purpose
    3. Predictive spark workflows support SSL

Predictive Workbench

New Features:
  1. UI has been redesigned for better performance and easier management.
    1. Separate workspace is provided for each Execution Engine-
      1. R
      2. Spark
      3. Python
      4. Java
    2. New and Improved look and feel –
      1. Progress Bar
      2. New icons – Run, Refresh, Parallel Processing
      3. Toast Messages
      4. Workflow panel with more comfortable provision to rename
      5. Upgraded to Angular JS 1.5.8
  2. Refactored and Optimized code –
    1. Improved integration and performance with R
    2. Introducing BizViz R package library
    3. Quicker PA services via dashboards  
    4. Incorporated BDB 3.5 Platform Optimization and refactoring factors
      1. OSGI Node redistribution and balancing
  3. Support for RJDBC is provided in R execution engine
  4. The following components support R Parallel Processing:
    1. Data Type Definition
    2. Filter
    3. Formula
    4. Naive Bayes with validation
  5. XG Boosting in R is enabled for Regression Analysis algorithm
  6. Cross-validation in R is allowed for Regression Analysis algorithm
  7. The following components are integrated with R Bokeh for interactive visualization:
    1. Linear Regression
    2. Multiple Linear Regression
    3. Logistic Regression
    4. Correlation
  8. Scheduler
    1. Introducing Python Workflows
    2. Users can now distribute the load by associating separate R servers while scheduling
  9. Data Writer- Introducing Python and R data writer components
  10. Data Reader- Introducing BizViz Data Store (Elastic Search) as a data reader component
  11. Introducing Queue mechanism for R workflows
  12. Python workflows can be deployed to Dashboard Designer to be consumed as data service/ data source
Enhancements:
  1. The Factor R Data Type has been added to the Data Type Definition component
Notes:
  1. Users must run a model in the workflow panel before saving it, else ‘File does not exist’ error message will be displayed while saving the model.
  2. Due to browser constraints, by default, only 50,000 cells can be viewed in the Result Tab Data Grid/Table. It can be configured through ‘start.r’ configuration file.
  3. Scheduling is not supported for Spark workflows.
  4. Python supports .CSV, MYSQL, MSSQL, BDB Data Store (Elastic Search) as a data source.
  5. Filter (for R Workflows)- In case of conditional statements error messages will be displayed if thrown by R, if R DF is saving all NA values, the error message will be empty.
  6. All Column/Field Headers do not support R keywords like NULL.
  7. ES Reader & Writer in R/Python Workspace – To execute the task at least one R/Python component should be present in the workflow
  8. Data Writer in Python Workspace - Byte Data Type is not supported.
  9. Scheduled jobs have not been classified according to the workspace types.
  10. R/Python workspace –Workflows with only Data Source and Data Writer components without any R/Python components will be saved under the ‘Data Preparation’ workspace.

iOS App

New Features:
  1. Server configuration is provided to connect any server and any account.
  2. Mobile-enabled Dashboards are integrated
  3. The following features are enabled in the Mobile Business Stories:
    1. Create New Business Stories from Data Stores
    2. Voice Assistance with the help of NLP
    3. Suggestions for voice-based queries
    4. View specific and Global filters are introduced
  4. User Documents:
    1. Users can access and use My Documents, Public Documents, and Favorites
    2. The document search is enabled through a ‘Search’ bar
    3. Documents re-arrangements and properties management for documents is possible now
Enhancements:
  1. The following functionalities are enhanced for ‘Documents’:
    1. Document creation
    2. Document Management
    3. Ambiguity suggestions
Recommendations:
Product Description
iOS App The application will run on iOS 10.3 or above versions
Business Story 1. Count and List operation can be performed only with one Dimension and N Filters
2. Top N/First N/Bottom N/Last N operations will display records according to N (If ‘N’ is not specified, it will take five records by default)
3. Top/First/Bottom/Last/Less/Greater should have one measure and one dimension and N number of filters
4. One dimension and one measure are required for Mixed chart
5. N number of dimensions and N number of measures are required for Grid
6. One measure, no dimension, and N filters are supported for Matrix Summary


Known Limitations:
Product Description
Business Story NLP does not support ‘and,’ ’or,’ ’not’ as the query suggestions.
Formula Field:
1.The dimension used in formula field creation cannot be used as filters or a merge conditions
2. NLQ does not work on a formula field
3. SQRT does not support negative values
4. Measure variable will support only double datatype
5. Dimensions created using formula editor cannot be used as slicers
6. Range feature does not support the ‘Sort’ functionality
Dashboard Designer
Custom ToolTip - Font Color properties do not support in IE.
Time-series chart animation does not work in IE browser.
Filters are available for the Data Store connector.
Undo and Redo functions do not work in Mobile, Tablet view, and while using the multi-select feature.
Multi-group-selection has some usability issues regarding synchronizing the selected components.
If selected components and dashboard both drag parallel, then it does not work smoothly. The functionality does not support in the Mobile View.
iOS App For creating a new story only one dimension and one measure combination is allowed
Toggling between charts for new stories is limited to 4
Data drilling inside charts can be achieved through filters