NVIDIA / NVIDIA/NeMo-Retriever
[FEA]: Add support for Kinetica as a VDB sink
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 3k
- Forks
- 349
- Avg merge
- 1d 23h
- Merged PRs (30d)
- 116
Description
Is this a new feature, an improvement, or a change to existing functionality?
New Feature
How would you describe the priority of this feature request
Currently preventing usage
Please provide a clear description of problem this feature solves
We at Kinetica would like to develop a feature which enables the nv-ingest services to use kinetica as a vdb sink. Right now, it seems the only sink available to use is Milvus. We would like to develop this feature so that kinetica is available as an alternative.
From the code https://github.com/NVIDIA/nv-ingest/blob/main/src/nv_ingest/modules/sinks/vdb_task_sink.py it seems there is no way to develop such a feature since it is tied to the Milvus libraries only. We are in the process of developing the Morpheus Vector Database Service for Kinetica as available in the issue https://github.com/nv-morpheus/Morpheus/issues/2058.
Describe the feature, and optionally a solution or implementation and any alternatives
Ideally, we are planning to develop a module like vdb_task_sink_kinetica.py which will enable Kinetica to be used as a vdb sink instead of Milvus.
Additional context
I think the module https://github.com/NVIDIA/nv-ingest/blob/main/src/nv_ingest/modules/sinks/vdb_task_sink.py needs to be altered suitably so that the Milvus specific code is factored out to a different module.
Any suggestions to how we may proceed would be extremely helpful.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading src/nv_ingest/modules/sinks/vdb_task_sink.py to understand how the current Milvus-specific sink is structured. Review the linked Morpheus issue about the Kinetica Vector Database Service and clarify the integration boundary before proposing the module or refactor. Done means Kinetica is available as an alternative VDB sink without leaving the existing Milvus path unusable.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- backend, databases
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 28/100