Separate names with a comma.
Discussion in 'iMac' started by atMac, Jul 1, 2011.
I'll just leave this here.
But you have 11.8GB unused RAM.
Mac OS X: Reading system memory usage in Activity Monitor
Making The Most of Activity Monitor
What Is Activity Monitor (or How to Take Your Mac's Pulse)
to ≠ too
People who actually believe $600 is the going rate for 16GB of RAM
That's why I paid under $200.
But you don't use it.
Still paid $100 too much
You beat me to it.
Unless you are heavy into image, film, video-game or music production or need to make some highly complex calculations, 16GB certainly would be a waste. 4GB-6GB is what 99% of the people use including gamers.
atMac, you make a good point about one of the least understood resources.
At 16GBs, you are still getting RAM shortages, as can be seen from the Page Outs, and although there is a lot of "inactive" RAM, you really have to understand that "inactive" means "not so recently used" as determined by OS X. Also, this pie diagram is just a point in time picture, and doesn't reflect the peaks. RAM demand varies by microseconds and the pie is updated once a second or less.
Thanks for the great example.
Only 16GB? Heh
I have the same problem as OP, but not this extreme.
Does anybody know how to release the Inactive memory to Free memory?
After 3 days of use, there were only 26mb of Free memory left out of 8Gb. But there were 1.4Gb of Inactive memory sitting around going nothing, while there were over 2Gb of Page out and Swap used. I had to reboot the machine to clear out the Inactive memory.
I don't see any advantages designing the OS to withhold vital system resources to a point the machine is forced to use free space on the hard drive as memory.
I could buy another 4Gb of RAM to make this happen less often, but as OP has posted, this does not solve this stupid design problem.
I can see high intensive rendering programs using whatever ram you can throw at it, if they can address it; but for my music and limited movie making stuff, 4 is quite enough. 4GB allows me to drive an entire orchestra quite comfortably in Logic Pro. Though to be fair, I wouldn't go less than that to do it.
My 13" MBP was the first machine, I bought that could actually do it. I'm just looking to see the size increase and price drop of SSD drives reach a comfortable fever pitch before I decide to swap out my 7200RPM drive I have in it now. That way I'll not only get a speed increase (due to the seek increase speed required to handle many many simultaneous disk reads at the same time (I do cue up to 15+ recordings at the same time), but I'll also gain the power savings that I voluntarily took a small hit on by upgrading to 7200RPM.
12gb for the win, not to much, and never to little
How did you get activity monitor to show how much (bytes/sec) in the Page in/outs?
I had long discussions about this issue and there are people who think it's a feature that os x blocks the ram with inactive ram till there is no free ram and it starts to write the hdd full with swap files and the system gets extremely slow...
Either reboot the system or open terminal and type "purge". This will free the inactive ram. The OS will freeze for some seconds and some processes will need to rewrite necessary data into the ram, but it helps.
I'm running Lion.
That's not how it works. Inactive, unmodified pages can be stolen any time.
Here is a description of how OS X paging works. Or actually a partial description
It is from http://developer.apple.com/library/...tual/ManagingMemory/Articles/AboutMemory.html
Google "Page Lists in the Kernel" (with the quotes)
The kernel maintains and queries three system-wide lists of physical memory pages:
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. In Mac OS X, 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.
The kernel moves pages from the active list to the inactive list if they are not accessed; it moves pages from the inactive list to the active list on a soft fault (see “Paging In Process”). When virtual pages are swapped out, the associated physical pages are placed in the free list. Also, when processes explicitly free memory, the kernel moves the affected pages to the free list.
Thanks for the info, it was an interesting read.
So OSX withhold recently used data in inactive memory in case its needed again, and when free memory become low, data in the inactive memory is dumped into hard drive before space are reallocated back to free memory to be used by systems.
This seems to me like complex solution to solve a simple problem. Why don't just dump the used data and free up ram like Windows.
This remind me a joke about when space race was on, the Americans spent vast amount of resources developing a pen that worked in zero gravity, while the Russians simply used pencils.
Ah awesome, thanks for clearing that up.
Im pretty sure the only way is to reboot, which is why i tell my friends not to leave there macs on for months, because they slow down.
there might be tools for emptying the inactive RAM, check google
Only the changed pages are Paged Out. They have to be paged out since they have been changed. Unchanged pages are just stolen from the inactive queue.
The objective is to optimize the use of resources.
The algorithm tries to defer the need to page out as long as possible, "hoping" that a changed page on the inactive queue will be reused, which will promote it to the active queue again. In this manner, we hope that page outs are minimized.
Every page out avoided represents a reduction in consumption of resources.
The article doesn't go into full detail, but it is possible that unchanged pages end up with a higher likelihood of being stolen, since there is a higher price to pay for stealing changed pages.
Sounds complex, but it all makes sense.
Inactive RAM is a disk cache that OS X uses to improve overall responsiveness.
You can see it work pretty easily:
1. reboot your Mac
2. open some app, e.g., Safari. Note the time it takes.
3. Quit safari
4. Launch Safari again.
Typically you'll see inactive memory go up after step 3, and you'll see that step 4 is a lot quicker than step 2 (especially with slower hard drives -- might not see any difference with an SSD. e.g.)
There are utilities to clear inactive RAM, but don't use them: they slow your system down by flushing the disk cache prematurely.
OP's chart doesn't necessarily show any problem at all. It shows that OS X has been able to use 12 GB of his RAM as a cache which is may be able to use to speed up his system.
This thread made me look at my System Memory usage, I originally had 12 GB and after looking at the screenshot of the OP I noticed a lot of Page Outs on my system.
I run VMWare Fusion and keep multiple Safari windows open along with Firefox (for netflix), iTunes (Radio), etc. Oddly, I noticed Safari had over a gig of memory in use but I never remember seeing my Free Memory going below ~3 gig so I thought I was good.
I recently purchased another 8 GB from Newegg.com (~$50 email special). I just installed the new memory and my Page outs are at 0 bytes after 1 day, I cant imagine this equates to much of a performance increase but it does indicate that 16 gig might be a sweet spot for my application use.
I will agree with most users that 12 GB might be the best bang for your $, but with memory going for so cheap I couldn't pass it up. The only concern I had was the sporadic lockups some users reported when updating to 16GB with their iMac. This was even more of a concern to me as I was mixing RAM manufactures. I haven't had any issue as of yet but will keep an eye on it.
Unless you're Steven Spielberg editing the sequel to Avatar...
Yes its massive overkill, even 8GB is too much...
There are many other applications and uses outside the film post production field which can benefit from more than 8GB of RAM.
Btw, editing film or video is not really RAM or CPU intensive, especially digital film, as it is using a very low compression and not AVCHD.
One could edit Avatar, or any other movie for that matter, on a G5 or even slower machine with the help of proxies (which are often used), but I guess with "edit" you mean all the fancy 3D stuff too, which can be created on low RAM machines too, as the rendering takes all the CPU and RAM power.