inactive RAM = 5GB?

firstyearprof · Feb 22, 2008, 11:05 AM

Hi,

I have a 2008 Mac Pro 3.0, 18GB RAM. From the "top" command in terminal, I'm showing 5GB of inactive RAM. I've read that such RAM is apple's way of keeping RAM in reserve in case the same application needs it in the future, but that it will give that RAM to another application as needed (or something like that).

HOWEVER, yesterday I was using an application that got up to about 12GB RAM, *and then it started swapping to hard drive* (the pageouts went from 0 to 327000), all the while that 5GB inactive RAM just sat right where it was.

I had been expecting my application to be able to access this inactive RAM once it ran out of free RAM rather than swapping to hard drive. What gives? I can't believe that the Leopard OS is this inefficient with memory allocation, so I'm guessing I'm not understanding something. Any ideas?

richthomas · Feb 22, 2008, 11:10 AM

im not sure but heres a good article from Apple that explains Mac OS X's memory management.

i think inactive RAM in only used if a new process requests it. so if your using the same RAM hungry app the inactive RAM wont be used.

BTW 18 GB! SWEEET! you should allocate some of it as a super fast scratch disk.

edesignuk · Feb 22, 2008, 11:16 AM

What was the application? Some applications themselves are only able to address certain amounts of memory.

kittiyut · Feb 22, 2008, 11:18 AM

Quote:

Originally Posted by firstyearprof

Hi,

I have a 2008 Mac Pro 3.0, 18GB RAM. From the "top" command in terminal, I'm showing 5GB of inactive RAM. I've read that such RAM is apple's way of keeping RAM in reserve in case the same application needs it in the future, but that it will give that RAM to another application as needed (or something like that).

HOWEVER, yesterday I was using an application that got up to about 12GB RAM, *and then it started swapping to hard drive* (the pageouts went from 0 to 327000), all the while that 5GB inactive RAM just sat right where it was.

I had been expecting my application to be able to access this inactive RAM once it ran out of free RAM rather than swapping to hard drive. What gives? I can't believe that the Leopard OS is this inefficient with memory allocation, so I'm guessing I'm not understanding something. Any ideas?

Which brand of RAM are you using ? (Hope not APPLE !!) And what are the temperature of each ? I noticed that on mine, DIMM 1 & 2 on the top riser are the hottest - at first I thought it was the RAM, but when I switch them to the bottom riser, the temp dropped and the replacement rose. I guess it is the position and/or being utilized more?

richthomas: how do set superfast scratch drive to use the extra RAM? I too have 18 GB and not fully utilizing them all.

richthomas · Feb 22, 2008, 11:21 AM

Quote:

Originally Posted by kittiyut

…richthomas: how do set superfast scratch drive to use the extra RAM? I too have 18 GB and not fully utilizing them all.

use this hint. you will need to mount it every time you want to use it.

firstyearprof · Feb 22, 2008, 11:23 AM

To answer:

The application is R (open source statistics) 64 bit. It should be able to access a virtually unlimited amount of RAM.

I have OWC RAM. My temps were a major concern (getting up to 176F !!) until recently - after putting them through a day of big time work, they now never get above 155F.

Anyway - I think this is a leopard issue... to wit: WHY IS LEOPARD ALLOWING INACTIVE MEMORY TO SIT THERE WHEN AN APPLICATION NEEDS IT?

kittiyut · Feb 22, 2008, 12:36 PM

Quote:

Originally Posted by richthomas

use this hint. you will need to mount it every time you want to use it.

Thanks for the link. However, when I try to mount the disk, I get permission denied???? Any suggestions?

firstyearprof · Feb 22, 2008, 01:45 PM

Hi all,

I think I understand what's going on. It is explained here:
http://developer.apple.com/documenta...outMemory.html

BEGIN QUOTE:

The active list contains pages that are currently mapped into memory and have been recently accessed.

The inactive list contains pages that are currently resident in physical memory but have not been accessed recently. These pages contain valid data but may be released from memory at any time.

The free list contains pages of physical memory that are not associated with any address space of VM object. These pages are available for immediate use by any process that needs them.

When the number of pages on the free list falls below a threshold (determined by the size of physical memory), the pager attempts to balance the queues. It does this by pulling pages from the inactive list. If a page has been accessed recently, it is reactivated and placed on the end of the active list. If an inactive page contains data that has not been written to the backing store recently, its contents must be paged out to disk before it can be placed on the free list. If an inactive page has not been modified and is not permanently resident (wired), it is stolen (any current virtual mappings to it are destroyed) and added to the free list. Once the free list size exceeds the target threshold, the pager rests.

END QUOTE

If I understand correctly, the page outs are not always a symptom of using virtual memory rather than RAM. Rather, when inactive memory is requested, it is given to the active application, but whatever is stored in inactive memory needs to be paged out to hard disk first - it is then given to the application.

So I guess that when my application requested more memory than was free, the OS first paged out the memory to the hard drive and then gave it to the application (it must have been a tiny amount given that I didn't notice the inactive memory decrease).

Does this make sense?

richthomas · Feb 24, 2008, 06:34 AM

Quote:

Originally Posted by kittiyut

Thanks for the link. However, when I try to mount the disk, I get permission denied???? Any suggestions?

use the disk number terminal quotes back after the first command. for example mine was disk4... yours maybe different.

Infrared · Feb 24, 2008, 08:51 AM

Quote:

Originally Posted by firstyearprof

If I understand correctly, the page outs are not always a symptom of using virtual memory rather than RAM. Rather, when inactive memory is requested, it is given to the active application, but whatever is stored in inactive memory needs to be paged out to hard disk first - it is then given to the application.

So I guess that when my application requested more memory than was free, the OS first paged out the memory to the hard drive and then gave it to the application (it must have been a tiny amount given that I didn't notice the inactive memory decrease).

Does this make sense?

Yeah, I think you got it more or less. What is happening to the real
memory used by your application (Real Memory column in Activity
Monitor and RSIZE column from 'top')? Is it hitting a 12GB ceiling?
I'd expect your memory-intensive application to be given pages stolen
from other less active applications (i.e., pages that were marked as
inactive and subsequently moved to the free pool).

Feb 22, 2008, 11:05 AM	#1
firstyearprof macrumors member Join Date: Apr 2007	inactive RAM = 5GB? Hi, I have a 2008 Mac Pro 3.0, 18GB RAM. From the "top" command in terminal, I'm showing 5GB of inactive RAM. I've read that such RAM is apple's way of keeping RAM in reserve in case the same application needs it in the future, but that it will give that RAM to another application as needed (or something like that). HOWEVER, yesterday I was using an application that got up to about 12GB RAM, and then it started swapping to hard drive (the pageouts went from 0 to 327000), all the while that 5GB inactive RAM just sat right where it was. I had been expecting my application to be able to access this inactive RAM once it ran out of free RAM rather than swapping to hard drive. What gives? I can't believe that the Leopard OS is this inefficient with memory allocation, so I'm guessing I'm not understanding something. Any ideas?

Feb 22, 2008, 11:10 AM	#2
richthomas macrumors 601 Join Date: Feb 2007 Location: /Aussieland/home	im not sure but heres a good article from Apple that explains Mac OS X's memory management. i think inactive RAM in only used if a new process requests it. so if your using the same RAM hungry app the inactive RAM wont be used. BTW 18 GB! SWEEET! you should allocate some of it as a super fast scratch disk. __________________ ♫Gee Officer Krupke, krup you!♫ 15" MacBook Pro Core Duo 2 GHz, 2 GB, 250 GB 7200.4; 23" Cinema Display; Snow Leopard Core i5 2.66 GHz, 4GB, 1.5 TB, 4870 1GB Nokia E63

Feb 22, 2008, 11:16 AM	#3
edesignuk Demi-God (Moderator emeritus) Join Date: Mar 2002 Location: Great(?) Britain	What was the application? Some applications themselves are only able to address certain amounts of memory. __________________ Search before you post: MRoogle **Better BBC Search**

Feb 22, 2008, 01:45 PM	#8
firstyearprof Thread Starter macrumors member Join Date: Apr 2007	I get it now I think Hi all, I think I understand what's going on. It is explained here: http://developer.apple.com/documenta...outMemory.html BEGIN QUOTE: The active list contains pages that are currently mapped into memory and have been recently accessed. The inactive list contains pages that are currently resident in physical memory but have not been accessed recently. These pages contain valid data but may be released from memory at any time. The free list contains pages of physical memory that are not associated with any address space of VM object. These pages are available for immediate use by any process that needs them. When the number of pages on the free list falls below a threshold (determined by the size of physical memory), the pager attempts to balance the queues. It does this by pulling pages from the inactive list. If a page has been accessed recently, it is reactivated and placed on the end of the active list. If an inactive page contains data that has not been written to the backing store recently, its contents must be paged out to disk before it can be placed on the free list. If an inactive page has not been modified and is not permanently resident (wired), it is stolen (any current virtual mappings to it are destroyed) and added to the free list. Once the free list size exceeds the target threshold, the pager rests. END QUOTE If I understand correctly, the page outs are not always a symptom of using virtual memory rather than RAM. Rather, when inactive memory is requested, it is given to the active application, but whatever is stored in inactive memory needs to be paged out to hard disk first - it is then given to the application. So I guess that when my application requested more memory than was free, the OS first paged out the memory to the hard drive and then gave it to the application (it must have been a tiny amount given that I didn't notice the inactive memory decrease). Does this make sense?

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Feb 22, 2008, 11:23 AM	#6
firstyearprof Thread Starter macrumors member Join Date: Apr 2007	To answer: The application is R (open source statistics) 64 bit. It should be able to access a virtually unlimited amount of RAM. I have OWC RAM. My temps were a major concern (getting up to 176F !!) until recently - after putting them through a day of big time work, they now never get above 155F. Anyway - I think this is a leopard issue... to wit: WHY IS LEOPARD ALLOWING INACTIVE MEMORY TO SIT THERE WHEN AN APPLICATION NEEDS IT?