有没有办法限制R进程的数量运行

Is there a way to limit the number of R processes running

我使用 doMC,它使用了包 multicore。当我调试(在控制台中)时,它发生了(好几次),它横向移动 fork-bombed.

R 有 setrlimit() 系统调用吗? 在 pyhton 中,我会使用 resource.RLIMIT_NPROC

理想情况下我想限制R进程的数量运行一个数字

编辑:OS 是 linux 分OS 6

您可以使用 registerDoMC(请参阅文档 here

registerDoMC(cores=<some number>)

另一种选择是在 运行 R 脚本之前使用 ulimit 命令:

ulimit -u <some number>

限制 R 能够生成的进程数。

如果你想限制多个 R 进程同时使用的 CPU 总数,你需要使用 cgroups or cpusets 并将 R 进程附加到 cgroup 或 cpuset。然后它们将被限制在 cgroup 或 cpuset 中定义的物理 CPUS。 cgroups 允许更多的控制(例如还有内存)但设置起来更复杂。

应该有几个选择。这是 Writing R Extensions, Section 1.2.1.1

的相关部分
   Packages are not standard-alone programs, and an R process could
contain more than one OpenMP-enabled package as well as other components
(for example, an optimized BLAS) making use of OpenMP. So careful
consideration needs to be given to resource usage.  OpenMP works with
parallel regions, and for most implementations the default is to use as
many threads as 'CPUs' for such regions.  Parallel regions can be
nested, although it is common to use only a single thread below the
first level.  The correctness of the detected number of 'CPUs' and the
assumption that the R process is entitled to use them all are both
dubious assumptions.  The best way to limit resources is to limit the
overall number of threads available to OpenMP in the R process: this can
be done via environment variable 'OMP_THREAD_LIMIT', where
implemented.(4)  Alternatively, the number of threads per region can be
limited by the environment variable 'OMP_NUM_THREADS' or API call
'omp_set_num_threads', or, better, for the regions in your code as part
of their specification.  E.g. R uses
     #pragma omp parallel for num_threads(nthreads) ...
That way you only control your own code and not that of other OpenMP
users.

我最喜欢的工具之一是控制它的包:RhpcBLASctl。这是它的描述:

Control the number of threads on 'BLAS' (Aka 'GotoBLAS', 'ACML' and 'MKL'). and possible to control the number of threads in 'OpenMP'. get a number of logical cores and physical cores if feasible.

毕竟您需要控制并行会话的数量以及分配给每个并行线程的 BLAS 核心数量。并行包默认每个会话 2 个线程是有原因的...

所有这些应该在很大程度上独立于 Linux 或 Unix 的风格 运行。好吧,除了 OS X 当然(仍然 !!)不会给你 OpenMP。

你可以从 doMC 和朋友那里控制最外层。