NVIDIA / NVIDIA/cuvs

[DOC] Enhance Memory Footprint Document

Open
#1,588 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

doc
Dominant language
Cuda
Stars
854
Forks
236
Avg merge
3d 3h
Merged PRs (30d)
62

Description

Add the memory footprint for the following workload:

  1. Streaming cagra index build -- build ivf-pq in batches and run cagra::optimize() [Relevant to Serverless deployment]
  2. Disk based cagra index build - building knn graph with data streaming from disk.
  3. CAGRA --> HNSW conversion

Document: https://docs.rapids.ai/api/cuvs/stable/neighbors/cagra/#memory-footprint

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the CAGRA Memory Footprint section in the linked documentation and review how existing workloads are described. Determine the documented memory footprint for streaming IVF-PQ/CAGRA builds, disk-based graph builds, and CAGRA-to-HNSW conversion; done means all three workloads are covered with clear, reproducible figures and deployment context.

Written by the indexing model from the issue text.

Assessment

Domain
documentation, performance
Issue type
Documentation
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.