BDB 4.0 release takes the platform into the league of world class E2E BI & Analytics platform. 4.0 is a Major release on top of 3.5 which took over 8 months. With this release we have completed our end to end Analytics platform story, the first team to build Data Pipeline, Data Wrangling, AI Workbench, Self Service BI, Search, Governed Dashboards and Mobility in a single product suite. All the products are General Available. Enterprise customers as well as Analytics partners can now use this product to build the new generation Analytics Platform for themselves or for their Vertical Segments. In 4.0 we have done a major Performance upgrade, where our dashboards are loading at 2X the speed, our search program searches through 120 million records in less than 5 secs.
Please see the performance report to understand what to expect from this platformClick here

Platform

Key Features:

1. Security enhancements: Introduced JWT token-based authentication to reduce the round trips, performance, and portability
2. Licensing control:
   1. Control the named users
   2. Control number of sessions per users
   3. License Expiry
3. Performance improvement on all platform services
4. Data Services performance improvement:
   1. Improved Parser to process the SQL statements
   2. Faster processing of SQL output
   3. Introduced user attribute-based data restrictions
5. Pluggable UI modules, this helps in deployment of individual modules
6. Banding and white-labeling container based on configuration files
7. New Data Connectors
   1. Data File Connectors
      1. Snowflakes
      2. PostgreSQL
      3. SAS
   2. API connectors
      1. ServiceNow
      2. Salesforce
      3. Facebook Ads
      4. Google AdWords
      5. Google Forms
      6. Twitter Ads
      7. LinkedIn Ads
      8. Firebase
      9. Leadsquared
8. Users can be created in bulk, by uploading user details in excel file.
9. Language setting for the dashboard, for multilingual dashboards

ETL

Key Features:

ETL with an integrated version of Data Preparation tool was introduced with an option to view, profile and cleanse data in a user-friendly interface. It focuses on non-technical customers. The users can perform the following tasks with the help of Data Preparation tool:

1. Displays the data paginated in a user-friendly grid.
2. Identifies the datatype, its type mismatches and computes format patterns.
3. Shows statistical information like count, distinct, duplicate for all columns.
4. 50 Transforms to correct and prepare the data for visualization.
5. View the list of transforms performed on a specific dataset.
6. The created Preparations can be exported to the following platform modules:
   1. ETL
   2. Data Pipeline
7. Users can save, edit, and re-export the edited preparation to another module using the Data Preparation tool.

Data Pipeline

Key Features:

The BDB Data Pipeline framework introduced with the following features:

1. Data Readers- SFTP, HDFS, Cassandra, JDBC (MySQL, MSSQL, Oracle, Postgre), Elastic Search, Mongo DB
2. Data Writers- HDFS, Cassandra, JDBC (MySQL, MSSQL, Oracle, Postgre), Elastic Search, Mongo DB
3. Transformation- Aggregation, Date Formatter, Split, Replace Text, Join, SQL Query
4. Model Runners- R, Spark, and Python models
5. Ingestion- SFTP Monitoring, Web Socket Listener, Sqoop Job
6. Custom Component Support- User can create their component using provided code structure.
   1. Scala Custom Component
   2. Python Custom Component
7. Job Deployment supported processes- Streamed and Invoked, Batch wise (job support)
8. Dynamic/Runtime update can be provided to the components
9. Data-lake: HDFS is supported with various file formats (such as CSV, Parquet, JSON, Avro)
10. PEM/PPK file support
11. Integration with the BDB Predictive models and Data Preparation scripts
12. Advanced Log displaying component deployment tracker
13. Excel Support for SFTP reader
14. Event sharing
15. File Splitter Component is introduced based on File Name, File Format, Regular-Expression (File-Name), Excel Sheet Name, Excel Sheet Index
16. Dynamic column header support for non-header Excel sheet.
17. Rule Splitter- Split data on conditions
18. Store Procedure Runner Component.
19. The user can select Pipeline deployment resource type configuration out of Low, Medium (default), and High

Predictive Workbench

Key Features:

1. New Workspaces added to the Predictive Workbench:
   1. Neural Network
   2. Auto ML
2. Tensor-board is added for NN Model training.
3. The ‘Predefined Scripts’ node provided for existing users and new users in the R and Python Workspaces with pre-packaged scripts.
4. Guided Tour implementation on the Landing page to introduce the major Predictive Workspaces.
5. R Workspace.
   1. Regression matrix in Performance.
6. New Algorithms added in the Python Workspace
   1. Naive Bayes Classifier component.
   2. Random Forest component
7. Apply Model implemented for Naive Bayes and Random Forest Algorithms in the Python workspace.
8. Zip Data source is available in R, Python, and NN workspaces.
9. User Interface option has been provided in the NN Workspace for NN model training.
10. PA Migration- supports migration of Predictive Workflows and Scripts have been provided.
11. CSV Reader and Data Service are completely moved to Python and R languages.
12. Elastic Search Reader and Writer added in R, Python, NN and Auto ML.
13. Changes done in the plugin services for performance improvement.
14. R Packages can be installed using the ‘Install Packages’ option provided on the Predictive Landing page.
15. Predictive models (R and Python) can be deployed to Data Pipeline.

Business Story

Key Features:

1. Enhanced AI-based Search: helps in faster data discovery .
2. Enhanced Formula Field Editor to support aggregated formulas.
3. New Charting Components:
   1. Pareto Chart
   2. Scatter Plot chart
4. Introduced Running summary as new aggregation .
5. Timeline Play: Visually run thru the changes that happened across timeline.
6. Guided tour introduced to the major sections of the Story plugin.
7. Websocket integration to create live Business Story.
8. Filter Enhancements:
   1. Support in the list and not in list operation
   2. Support like operator for dimensions
9. Automated Insights: System will identify the most relevant measures and dimensions and automatically build visualization, which user can pin to his Business story.

Dashboard Designer

Key Features:

1. AI-based Search component for quick and responsive data search and visualization.
2. Leaflet Map component provided with clustering and color filling on regions and markers.
3. Custom Chart component with 5 pre-configured D3 charts to integrate any third party visual.
4. FAN Chart for plotting uncertain data.
5. Mito Plot chart provided for genome study.
6. Action Menu on Charts to change the view of your chart at run time. 
7. Language mapping on dashboards.
8. Custom Tooltip with improved mobile tooltip and more configuration options.
9. Guided Tour introduced to Dashboard Designer.
10. Web Socket data connector
11. Predictive Connector with Python workflows support.
12. ‘Dark’ and ‘Material’ themes introduced for dashboards.
13. ‘Help’ section with detailed scripting provided for the advanced users.

Mobile Apps

Key Features:

1. Native mobile app introduced with ‘Show Story’ and ‘Show Dashboard’ functionalities.
2. ‘Create Story’ functionality was provided with the following features:
   1. Supported Charting Components: Column, Line, Bar, Pie charts.
   2. AI-based Search with following enhancements.
      1. Search bar supports text-based search
      2. Displays the saved Search Queries
      3. Displays Dimensions and Measures information
      4. Color-coded display of the searched Dimensions and Measures in the result view
      5. Expand Search Result view
   3. Filter Panel provided with ‘In’ and ‘Not-in’ options to apply filter conditions on data
   4. Secured ‘My Account’ configuration functionality introduced with the following options:
      1. Change Password
      2. View API token
   5. Document Properties to share Story documents with individual users and user groups.