Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature Request]: OpenSearch Byte Vector Support #17271

Open
christopher-learningpool opened this issue Dec 13, 2024 · 0 comments
Open

[Feature Request]: OpenSearch Byte Vector Support #17271

christopher-learningpool opened this issue Dec 13, 2024 · 0 comments
Labels
enhancement New feature or request triage Issue needs to be triaged/prioritized

Comments

@christopher-learningpool

Feature Description

Add support for byte vector storage in the OpenSearch vector store integration, leveraging OpenSearch 2.17's new byte vector capabilities with the Faiss engine. This would allow users to store vectors more efficiently by using 8-bit integers (-128 to 127) instead of floats, significantly reducing storage requirements while maintaining search quality.

Reason

Currently, the OpenSearch integration in LlamaIndex only supports float vectors. Adding byte vector support would require:

  • Adding a new data_type parameter to OpensearchVectorClient
  • Updating the index creation and query logic to handle byte vectors

Value of Feature

Storage Efficiency

  • Byte vectors require 1/4 the storage space of 32-bit float vectors
  • Reduced storage costs for large-scale deployments
  • Faster network transfer for distributed systems

Performance Benefits

  • Potentially faster similarity search due to reduced memory bandwidth requirements
  • More efficient cache utilisation
  • Better scalability for large vector databases
@christopher-learningpool christopher-learningpool added enhancement New feature or request triage Issue needs to be triaged/prioritized labels Dec 13, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request triage Issue needs to be triaged/prioritized
Projects
None yet
Development

No branches or pull requests

1 participant