rapidsai/cudf

Use grid stride in CSV reader kernels

Open

#14,066 创建于 2023年9月8日

在 GitHub 查看
 (1 评论) (0 反应) (1 负责人)C++ (6,000 star) (735 fork)batch import
PerformancecuIOgood first issue

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.

贡献者指南

技术栈
cpp
领域
backenddataperformance
议题类型
performance
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
4
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
half day
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
blocked
清晰度议题是否清楚说明期望改动、验收标准和下一步。
clear
前置要求
CUDA programminggrid stride loopsC++
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
20
研究方向
Investigate the current CSV reader kernel implementations in the cudf repository, specifically the parser and data inference kernels. Look for where threads are assigned per row and refactor to use a grid stride loop. This will involve modifying CUDA kernel launches and loop structures in files such as csv reader.cu or similar. Refer to existing CUDA grid stride examples in the codebase for pattern consistency.