admin管理员组文章数量:1122846
I have a 12GB feather file that I read into R with
data_df <- arrow::read_feather('fname.feather')
It often takes around 2 seconds. But sometimes it can take around 60 seconds. I have noticed it being this slow in RStudio IDE and Positron IDE, but also in the R console in the system terminal, so it does appear to be the R interface that is the issue.
I can not find a way to reproduce it but have experienced it repeatedly for many months.
Does anyone know of any reason why things must take 2 seconds usually, but be painfully slow sometimes, when reading feather files?
I have 96GB of RAM and most of it is free when I read in the feather file.
- My system is Arch Linux
- CPU: AMD Ryzen 9 7900X 12-Core, 96GB DDR5 RAM.
- R is 4.4.2; arrow package is 17.0.0.1
- RStudio 2024.09.1.394-1 , Positron 2024.11.0.116-1
本文标签: High variability in reading speeds of feather with arrowreadfeather in RStack Overflow
版权声明:本文标题:High variability in reading speeds of feather with arrow::read_feather in R - Stack Overflow 内容由网友自发贡献,该文观点仅代表作者本人, 转载请联系作者并注明出处:http://www.betaflare.com/web/1736300739a1930927.html, 本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如发现本站有涉嫌抄袭侵权/违法违规的内容,一经查实,本站将立刻删除。
发表评论