-
Notifications
You must be signed in to change notification settings - Fork 903
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
Add compute_shared_memory_aggs used by shared memory groupby #17162
Add compute_shared_memory_aggs used by shared memory groupby #17162
Conversation
Co-authored-by: David Wendt <[email protected]>
@davidwendt would you please take another pass over this PR? |
@@ -23,6 +23,91 @@ | |||
#include <cuco/static_set_ref.cuh> | |||
|
|||
namespace cudf::groupby::detail::hash { | |||
// TODO: TO BE REMOVED issue tracked via #17171 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this still relevant?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, #17171 should be addressed to avoid code duplication.
Co-authored-by: David Wendt <[email protected]> Co-authored-by: Nghia Truong <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Approving CMake. A couple tiny C++ suggestions (non-blocking).
Co-authored-by: Bradley Dice <[email protected]>
/merge |
Description
This work is part of splitting the original bulk shared memory groupby PR #16619.
This PR introduces the
compute_shared_memory_aggs
API, which is utilized by the shared memory groupby. The shared memory groupby process consists of two main steps. The first step was introduced in #17147, and this PR implements the second step, where the actual aggregations are performed based on the offsets from the first step. Each thread block is designed to handle up to 128 unique keys. If this limit is exceeded, there won't be enough space to store temporary aggregation results in shared memory, so a flag is set to indicate that follow-up global memory aggregations are needed to complete the remaining aggregation requests.Checklist