July 5th, 2018
Platform
Fixes:
- Dashboards are migrated successfully now (BP-2504).
- (Audit Trail) If the audit settings are disabled, data is not written to DB now (BP-2444)
- (Auto Schedule) The newly created auto-scheduled/refreshed datastores are now listing in the ‘Schedule Monitoring’ (BP-2412)
- (Data Center) Able to create Spark SQL Data Set now (BP-2372)
- (Data Center) Datastore refresh request gets delivered to AMQ for specific datastores (BP-2369)
- (Data Center) After uploading a file and creating datastore based on the same file, the file data gets written to the elastic search engine (BP-2356)
- (Data Center) Able to create MSSQL Data Set now (BP-2339)
- (Data Center) Data Restriction works properly in Data Store (BP-1977)
- (Data Center) Users can flawlessly create ‘Drill Definition’ in the Datastore Meta Data (BP-1779)
- (Data Center) Latest Schedule Status properly displays failure/success message (BP-2348)
- (Google Sheet API) Decimal Numbers are displayed properly now (BP-2398)
- (MailChimp API) Users can flawlessly create MailChimp Data Connector now (BP-2346)
- (Google Analytics API) Dynamic Date Type works properly while creating a Google Analytics Data Set (BP-2344)
- (Home UI) Users cannot open a document from the open doc with an old API Token. (BP=2351)
- (Forgot Password) Users assigned to the Admin Role can reset the password using the ‘Forgot Password’ option (BP 1029)
- (Windows AD Authentication) Windows AD ‘User Groups’ lists under Sync configuration, after configuring the same from the ‘Admin’ module (BP-2347)
- (Play) The ‘Exports to PPT’ option works flawlessly for the Play plugin (BP-2361)
Data Preparation
Fixes:
- Nulls out the variables and object references for the garbage collector to resolve the memory issue (DAPR-585)
Business Story
Fixes:
- (Sort) Manual sort works flawlessly now (BP-2349)
- (Filter) Users can successfully apply a combination of the global and local filter (BP-2340)
- (Alert Center) Alert triggers correctly (BP-2355)
Survey
Fixes:
- (Data Mart) Users can successfully create Data Mart in Survey (BP-2368)
Dashboard Designer
Charting
Enhancements:
- (SDK Methods) SDK methods are updated for handling exceptions (BDD-380)
Data Pipeline
New Features:
- The BDB Data Pipeline framework supports the following
list of features:
- SFTP
- S3
- HDFS
- Cassandra
- JDBC (MYSQL, MSSQL, ORACLE, POSTGRE)
- Elastic Search
- Data Writers
- S3
- HDFS
- Cassandra
- JDBC (MYSQL, MSSQL, ORACLE, POSTGRE)
- Elastic Search
- Transformation
- Aggregation
- Date Formatter
- Date Formatter
- Replace Text
- Join
- ML Model Runners
- R Model
- Spark Model
- Ingestion:
- SFTP Monitoring
- Web Socket Listener
- API Ingestion
- Web Socket Broadcast
- Custom Component Support
- Job Deployment Processes/Type
- Streamed
- Scheduled
- Invoked
- Dynamic Component Update
- Data-lake:
- At present HDFS and S3 are supported with various file formats such as CSV, Parquet, JSON, Avro
Predictive Workbench
Enhancements
- Predictive Settings are updated in the Admin panel (BPR-840)
Fixes:
- (Data Preparation) Column converted with Data Type Definition works for ‘Operator’ option in R Formula (BPR-1460)
- (R Servers) Multiple R servers can be configured in Scheduler (BPR-1467)
- (Summary) Component summary now displays from the left side (BPR-1476)
- (Algorithms) Enhancement is done on the ‘Summary’ of the R Decision Tree (BPR-1477)
- (Data Preparation) Missing Values replacement can work with Outlier algorithm (BPR-1478)
- (Data Source) Enhancement is done on advanced SQL queries (BPR-1480)
- (Data Writer) The Internal Data Writer now supports Saved Models of Data Type Definition (a Data Preparation component), Regression, Forecasting (R Double Exponential Smoothing), Classification, Association, and Outliers (BPR-1483)
- (Algorithm) Number of outlier values change by changing fence co-efficient in Outlier (BPR-1493)
- (Data Preparation) R Normalization displays proper values for the ‘Zero-Score’ option (BPR-1511)
- (Data Preparation) The ‘Result’ tab displays proper column headers for R Formula (BPR-1530)
- (Live Job Status) A warning message displays while deleting a job (BPR-1538)
- (Algorithm) XG Boosting of Regression algorithm works with Apply Model component (BPR-1548)
- (Data Writer) Data Store Writer supports parallel workflow (BPR-1553)
- (Algorithm) The ‘Result’ of Association algorithm is realigned (BPR-1502)
Recommendations:
Product | Description |
---|---|
Data Pipeline | |
Following are the basic requirements to deploy Data Pipeline on-premises | |
1. Kubernetes Master – 8 GB RAM 4 Core | |
2. Kubernetes Slave – 32 GB RAM 4 Core (Min: 3 Slaves) | |
3. Kubernetes 1.9 or Higher |
Known Limitations:
Product | Description |
---|---|
Predictive Workbench | |
The ‘Share’ option does not work for predictive workflows without selecting Admin role and Viewer role (BPR-1731) | |
In R and JAVA workspaces, MSSQL writer does not support special characters. (BPR-1728) | |
The parallel process supports up to 2 threads in concurrent (BPR-1661) | |
Workflow does not support any other algorithm after Apply Model if Saved Model is connected to that Apply Model component (BPR-1611) | |
(Data Preparation) The ‘String’ column data type lists as ‘Double’ for the Data Type Definition saved model (BPR-1545) | |
(Live Job Status) The ‘Status’ tab becomes blank while clicking the ‘Refresh’ button (BPR-1537) | |
(R Workspace) Data with the format YYYY/MM/DD and DD/MM/YYYY are not supported in R (BPR-1301) | |
(Data Preparation) The ‘BLANK’ option does not work for the R Formula component (BPR-511) |