Skip to content

Multi-GPU Batched KMeans #2017

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.

Sign up for GitHub

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

Open
viclafargue wants to merge 5 commits into
base: main
Choose a base branch
from
Open

Multi-GPU Batched KMeans #2017

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
07707af
Multi-GPU Batched KMeans
viclafargue Apr 13, 2026
efc270f
Merge branch 'main' into mg-batched-kmeans
viclafargue Apr 13, 2026
7197b71
cluster_cost on device
viclafargue Apr 14, 2026
84ab315
Updated testing
viclafargue Apr 14, 2026
47d4b94
templating
viclafargue Apr 15, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 9 additions & 1 deletion cpp/include/cuvs/cluster/kmeans.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -167,6 +167,13 @@ enum class kmeans_type { KMeans = 0, KMeansBalanced = 1 };
* on the host. Data is processed in GPU-sized batches, streaming from host to device.
* The batch size is controlled by params.streaming_batch_size.
*
* Multi-GPU dispatch is selected automatically based on the handle state:
* - If `raft::resource::is_multi_gpu(handle)` (cuVS SNMG): the full dataset X
* is split across GPUs internally with an OpenMP parallel region and NCCL.
* - If `raft::resource::comms_initialized(handle)` (Dask/Ray): X is treated as
* this worker's partition, and RAFT communicators are used for collectives.
* - Otherwise: single-GPU batched k-means.
*
* @code{.cpp}
* #include <raft/core/resources.hpp>
* #include <cuvs/cluster/kmeans.hpp>
Expand Down Expand Up @@ -196,7 +203,8 @@ enum class kmeans_type { KMeans = 0, KMeansBalanced = 1 };
* raft::make_host_scalar_view(&n_iter));
* @endcode
*
* @param[in] handle The raft handle.
* @param[in] handle The raft handle. When a multi-GPU resource is
* attached, multi-GPU dispatch is used automatically.
* @param[in] params Parameters for KMeans model. Batch size is read from
* params.streaming_batch_size.
* @param[in] X Training instances on HOST memory. The data must
Expand Down
Loading
Loading