Add Vector Search workload to benchmark #9141
Conversation
Signed-off-by: Archer <[email protected]>
Naarcha-AWS
requested review from
kolchfa-aws,
AMoo-Miki,
natebower,
dlvenable and
epugh
as code owners
|
Thank you for submitting your PR. The PR states are In progress (or Draft) -> Tech review -> Doc review -> Editorial review -> Merged. Before you submit your PR for doc review, make sure the content is technically accurate. If you need help finding a tech reviewer, tag a maintainer. When you're ready for doc review, tag the assignee of this PR. The doc reviewer may push edits to the PR directly or leave comments and editorial suggestions for you to address (let us know in a comment if you have a preference). The doc reviewer will arrange for an editorial review. |
|
@OVI3D0: I think cluster requirements would be a great addition. That being said, are the cluster requirements general for all workloads, or do we have specific cluster requirements/recommendations for each? This is meant to be more of a general overview, but any additional detail/guidance we can add would be helpful. |
_benchmark/workloads/vectorsearch.md
Outdated
|
|
||
| # Vector search | ||
|
|
||
| The vector Search workload benchmarks OpenSearch's Vector Engine capabilities for both indexing and search operations. It tests various vector search algorithms, quantization methods, and index configurations to measure performance metrics like throughput, latency, and recall accuracy. The workload supports different datasets and can evaluate both trained and untrained vector search methods. |
There was a problem hiding this comment.
Search -> search in the first sentence
There was a problem hiding this comment.
This looks fine to me. We can have this merged in for now and get feedback on it from users.
As discussed in our previous meeting, this is mostly identical to the Vectorsearch README from the workloads repository. The concern is if the README evolves over time, we'll need to update this documentation accordingly. If we plan to differentiate this documentation from the workload's README, we should consider either simplifying this or making this more of an intuitive "practical" guide instead. That way, this documentation can stand on its own and we can still link the README if users need more technical support.
One thing we could do is give more examples about how to/when to incorporate certain workload-params. |
Signed-off-by: Naarcha-AWS <[email protected]>
Checklist
For more information on following Developer Certificate of Origin and signing off your commits, please check here.