rapidsai/cudf

Use grid stride in CSV reader kernels

開放

#14,066 建立於 2023年9月8日

 (1 則留言) (0 個反應) (1 位負責人)C++ (735 個分叉)batch import
PerformancecuIOgood first issue

倉庫指標

星標
 (6,000 顆星)
PR 合併指標
 (平均合併 17天 21小時) (30 天內合併 230 個 PR)

描述

Currently, the CSV reader parses data using a thread per row, and a separate thread is used for each row, regardless of the file size. Using a grid stride loop would allow kernels to launch with preset number of blocks even with large input.

This applies both to the parser and the data inference kernels.

貢獻者指南