Content Comparison

• BulkPM API fix
• New remote debugging feature
• Remove DMaaP MR dependency
• Support new JanusGraph version (TBD)

1. provide incident API

1. Preliminarily provide an incident management framework to reduce the pressure on OSS/BSS by aggregating lots of anomalous data into a small number of incidents. 

1. Continue Performance Improvements
2. (new) VES events for CM Changes
3. Delta/Comparison feature
4. Model Upgrade using 'ModuleSetTag' for better efficiency & performance
5. Subscription based CM Data Notifications
6. Support for alternate (3GGP FDN) identifier

1. Faster processing of large requests
2. Message based integration for 3PP interested in CM updates
3. Compare current CM data with original to help with optimization, & troubleshooting
4. More efficient and faster upgrade for many (similar) nodes
5. Finer control of notifications ie. control only relevant data changes are notified
6. 3GPP Compliance, prepare for 3GPP compliant interfaces

• VES Collector
  • remove DMaaP MR dependency
• SNMP Trap Collector
  • Updates of outdated libraries

No changes

OSA: no changes as there were no vulnerabilities raised externally to ONAP 

SECCOM:

• Recommendations for package upgrades
• Add python compatibility module
• CLM scan for Portal NG

DMaaP removed from policy framework

Tracing support added in policy CLAMP ACM

Support for outProperties in restart and redeploy scenarios of CLAMP ACM.

Backward compatibility support added for ACM participants

Regression test suite added for CLAMP ACM

• Amount of security vulnerabilities drastically reduced
• Debugging and diagnostics in CLAMP ACM is enhanced with tracing support
• Users can retain outProperties history of elements after redeployment.
• Users can use the older version of participants against the new ACM-R releases with optional upgrade to latest features.

• use non-root user in containers
• update dependencies

1.Provide generative AI interactive functionality based on large models.
2.Provide the function of storing usage records of large models
3.Provide large model evaluation functionality to assess different large models.

1.Based on user input, use a large model to identify and analyze user intentions, and provide feedback information.
2.Learn and analyze user historical data as training data for large models.
3.Evaluate the performance of different large models and assist users in selecting the optimal large model.