I run parallel dataframe processing on Windows 10 in the fashion:
(我以以下方式在Windows 10上运行并行数据帧处理:)
Parallel(n_jobs=28)(
delayed(function)(group) for name, group in grouped_data)
SMT is disabled in Ryzen-Master and I see only 32 physical cores in any monitoring software.
(在Ryzen-Master中禁用了SMT,并且在任何监视软件中我都只能看到32个物理内核。)
When the script is being executed I see only 16 cores loaded 80-90%, and all the other 16 cores idle. (执行脚本时,我看到只有16个内核加载了80-90%,其他所有16个内核都处于空闲状态。)
If I enable SMT (32 physical cores and 64 logical) and launch it with (如果启用SMT(32个物理核心和64个逻辑核心)并使用)
n_jobs = 60
then I see the same picture: only fist 16 physical and first 32 logical cores are loaded, others idle.
(然后我看到的是同一张图片:仅加载了16个物理内核和前32个逻辑内核,其他空闲。)
Because some business limitations I can't install and check it on Linux.
(由于某些业务限制,我无法在Linux上安装和检查它。)
What is the problem? (问题是什么?)
ask by Иван Судос translate from so 与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…