Details
-
New Feature
-
Status: Resolved
-
Major
-
Resolution: Fixed
-
None
-
None
Description
The purpose would be to write only the record batch to the device memory, while leaving the Schema on the CPU. Similarly, record batches can be reconstructed through pointer arithmetic (the IPC loading as in arrow/ipc/reader.h) without having to copy the data to main memory
Attachments
Issue Links
- is part of
-
ARROW-1055 [C++] GPU support library development
- Resolved