3

I have a question regarding R and RStudio on my Windows Server 2012 R2 Datacenter.

I access the server directly and the server has 128GB RAM. But when I run my R-script, the task manager shows me, R is only using up to 2 MB of RAM and an overall of 4% of the RAM of the whole system are used.

I have the current R-Version installed and am using the 64-System option in RStudio. When I check my memory memory.limit() R says, that I have 1.759219e+13 of memory, which is not possible. I can't change the memory-limit or anything else.

Whenever I acess R directly through Rgui and type in the lines, R tells me the following

> memory.limit()
[1] 131023
> memory.size()
[1] 31.5

Whenever I use R on my Laptop, it uses about 5GB of RAM. So I am wondering, what is happening here.

Questions:

1.) Can I use R and RStudio on a Windows 2012 Server (I feel as if that should be okay, although I didn't find a lot of information on that on the internet)

2.) How can I provide the R-process with more memory, so the computation will be faster?

Thank you very much for your help. As this is my first question, let me please know if you need any additional information.

Edit

Maybe this helps explaining what my problem is. Why can't the R-process get more power?

Resource-Monitor and Task-Manager:

Resource-Monitor and Task-Manager]

CaldeiraG
  • 2,623
  • 8
  • 21
  • 34

1 Answers1

0

so I figured some things out and wanted to share it with you.

First, the Information about the wrong memory "memory.limit() R says, that I have 1.759219e+13" is due to the current R-build. I downgraded to the previous version and didn't get the error.

Second, it is perfectly fine to install the RStudio Desktop-Version on the server. The only drawback is, you have to sign in your server directly/with remote desktop control. Other than that, it works as regulare desktop with better hardware.

Third, by nature, R is designed to only use one core for each R instance. Apparently, you can either use special R functions for simultaneous computation (parallel), or you just start multiple R instances (having multiple R sessions open), so each session is using a different core. With different R instances open, I can use all my RAM. You just have to break down your code.

I posted this answer in case, anybody else would have these problems. I hope it is the correct way to post an answer and not to edit my post. Please let me know if I should correct this.

Thanks