Comments on: Hello Dynamic Memory? http://www.vcritical.com/2010/02/hello-dynamic-memory/ Informed Virtualization Criticism Fri, 30 Jul 2010 12:54:25 -0700 http://wordpress.org/?v=2.9.1 hourly 1 By: Tweets that mention Post-R2 release of Hyper-V may include memory overcommit - Dynamic Memory | VCritical -- Topsy.com http://www.vcritical.com/2010/02/hello-dynamic-memory/#comment-9655 Tweets that mention Post-R2 release of Hyper-V may include memory overcommit - Dynamic Memory | VCritical -- Topsy.com Sat, 06 Feb 2010 16:30:27 +0000 http://www.vcritical.com/?p=2270#comment-9655 [...] This post was mentioned on Twitter by Eric Gray, Eric Gray, Eric Gray, William Lam, Andrew Storrs and others. Andrew Storrs said: RT @eric_gray: VCritical: Hello Dynamic Memory? http://bit.ly/91tAAs [...] [...] This post was mentioned on Twitter by Eric Gray, Eric Gray, Eric Gray, William Lam, Andrew Storrs and others. Andrew Storrs said: RT @eric_gray: VCritical: Hello Dynamic Memory? http://bit.ly/91tAAs [...]

]]> By: Phil http://www.vcritical.com/2010/02/hello-dynamic-memory/#comment-9653 Phil Fri, 05 Feb 2010 16:56:44 +0000 http://www.vcritical.com/?p=2270#comment-9653 Mark: "That’s where VMware DRS comes in…" or PROTips in a Hyper-V environment. Mark: “That’s where VMware DRS comes in…”

or PROTips in a Hyper-V environment.

]]> By: Fernando http://www.vcritical.com/2010/02/hello-dynamic-memory/#comment-9650 Fernando Thu, 04 Feb 2010 10:31:14 +0000 http://www.vcritical.com/?p=2270#comment-9650 Brilliant explanation :-D Brilliant explanation :-D

]]> By: Mark Chuang http://www.vcritical.com/2010/02/hello-dynamic-memory/#comment-9649 Mark Chuang Wed, 03 Feb 2010 20:08:26 +0000 http://www.vcritical.com/?p=2270#comment-9649 @James It feels like saying that "VMware's approach can lead to paging" while "MS's approach won't" misses the bigger picture. Let's look at the scenario that would lead to paging for VMware, i.e. the VMs on a host cumulatively need more memory than exists on the host. In the MS scheme (if I'm interpreting your explanation correctly), the MS balloon driver will try to free up memory, but it won't be able to because all of the memory is actively being used. So you end up with Hyper-V VMs that need more memory but can't get it, so the performance for those VMs are essentially capped as well. So in both approaches, the bigger picture/solution is that you'll need to move 1 or more VMs to another host (since AFAIK, neither company has figured out how to defy the laws of physics and create more physical memory out of thin air). That's where VMware DRS comes in... (Disclosure: Yes, I work for VMware.) @James

It feels like saying that “VMware’s approach can lead to paging” while “MS’s approach won’t” misses the bigger picture.

Let’s look at the scenario that would lead to paging for VMware, i.e. the VMs on a host cumulatively need more memory than exists on the host. In the MS scheme (if I’m interpreting your explanation correctly), the MS balloon driver will try to free up memory, but it won’t be able to because all of the memory is actively being used. So you end up with Hyper-V VMs that need more memory but can’t get it, so the performance for those VMs are essentially capped as well.

So in both approaches, the bigger picture/solution is that you’ll need to move 1 or more VMs to another host (since AFAIK, neither company has figured out how to defy the laws of physics and create more physical memory out of thin air). That’s where VMware DRS comes in…

(Disclosure: Yes, I work for VMware.)

]]> By: Anton Zhbankov http://www.vcritical.com/2010/02/hello-dynamic-memory/#comment-9648 Anton Zhbankov Wed, 03 Feb 2010 18:10:29 +0000 http://www.vcritical.com/?p=2270#comment-9648 James, I know how can you give VM more memory - Windows 2003 Enterprise and above support memory hot-add. But how can you hot remove memory? I see only one way to do this - ballooning = swapping. James, I know how can you give VM more memory – Windows 2003 Enterprise and above support memory hot-add. But how can you hot remove memory?
I see only one way to do this – ballooning = swapping.

]]> By: Anton Zhbankov http://www.vcritical.com/2010/02/hello-dynamic-memory/#comment-9647 Anton Zhbankov Wed, 03 Feb 2010 18:06:56 +0000 http://www.vcritical.com/?p=2270#comment-9647 It's dynamic memory! ;) It’s dynamic memory! ;)

]]> By: James O'Neill http://www.vcritical.com/2010/02/hello-dynamic-memory/#comment-9646 James O'Neill Wed, 03 Feb 2010 17:21:47 +0000 http://www.vcritical.com/?p=2270#comment-9646 Anton, you're right we will probably see lots of arguments about the correct usage of "commit" and "Overcommit". Overcommitment (at least when I was introduced to the term in the context of non-virtualized OSes) is the point where paging begins. Far be it from me to advise VMware on how to market their products but it would seem to me that the whole upside on page sharing is that it allows (sum of granted) > (memory physical) without actually being overcommitted and needing to page (bad). Microsoft's dynamic memory (as seen in 2008) didn't page - so it reduced what was granted to one VM to increase what was granted to another. Different beasts. Anton, you’re right we will probably see lots of arguments about the correct usage of “commit” and “Overcommit”. Overcommitment (at least when I was introduced to the term in the context of non-virtualized OSes) is the point where paging begins.

Far be it from me to advise VMware on how to market their products but it would seem to me that the whole upside on page sharing is that it allows (sum of granted) > (memory physical) without actually being overcommitted and needing to page (bad).

Microsoft’s dynamic memory (as seen in 2008) didn’t page – so it reduced what was granted to one VM to increase what was granted to another. Different beasts.

]]> By: Fernando http://www.vcritical.com/2010/02/hello-dynamic-memory/#comment-9644 Fernando Tue, 02 Feb 2010 22:39:26 +0000 http://www.vcritical.com/?p=2270#comment-9644 So, what is it then ? So, what is it then ?

]]> By: Anton Zhbankov http://www.vcritical.com/2010/02/hello-dynamic-memory/#comment-9638 Anton Zhbankov Tue, 02 Feb 2010 16:47:59 +0000 http://www.vcritical.com/?p=2270#comment-9638 James, AFAIR VMware defines Memory Overcommitment as (Sum of memory granted) > (Memory physical). So if we have 32GB of memory and 40GB granted to running VMs - memory is overcommited, even if working set is 15GB. TSP is included to Memory Overcommitment techniques with ballooning. So actually it's just a question of terms. Marketing always will find something to say, no matter what engineers do. James,
AFAIR VMware defines Memory Overcommitment as (Sum of memory granted) > (Memory physical). So if we have 32GB of memory and 40GB granted to running VMs – memory is overcommited, even if working set is 15GB.

TSP is included to Memory Overcommitment techniques with ballooning. So actually it’s just a question of terms.

Marketing always will find something to say, no matter what engineers do.

]]> By: James O'Neill http://www.vcritical.com/2010/02/hello-dynamic-memory/#comment-9637 James O'Neill Tue, 02 Feb 2010 15:17:36 +0000 http://www.vcritical.com/?p=2270#comment-9637 @Anton. Overcommitment means: Sum of working Set > Total available memory --> Paging --> poor perf. When VMware products share pages Sum of Allocated Memory > total memory > Sum of working set That's not overcommitment: Working Set no paging --> Good perf Unfortunately people don't always distinguish. The scheme Mark describes had dynamic allocation so Sum of Max memory > Total memory Sum of allocated memory < Total memory In other words once all memory is allocated , the memory available to one VM can only be increased is if the memory available in another is reduced by a balloon driver. [Eric quoted Mark's explanation of that - just follow the link in the original post above.] If this technology reaches the market as it was when Mark saw it (and it would be a brave person to extrapolate from a single screen shot to a release schedule), VMware folks will stop saying "Microsoft is copying what we do" and start saying "Microsoft doesn't do all the things we do" and the argument will move to how good or bad paging in the Virtualization layer is really (VMware do it, Microsoft don't). @Anton.
Overcommitment means:
Sum of working Set > Total available memory –> Paging –> poor perf.
When VMware products share pages
Sum of Allocated Memory > total memory > Sum of working set
That’s not overcommitment: Working Set no paging –> Good perf

Unfortunately people don’t always distinguish.

The scheme Mark describes had dynamic allocation so
Sum of Max memory > Total memory
Sum of allocated memory < Total memory

In other words once all memory is allocated , the memory available to one VM can only be increased is if the memory available in another is reduced by a balloon driver. [Eric quoted Mark's explanation of that - just follow the link in the original post above.]

If this technology reaches the market as it was when Mark saw it (and it would be a brave person to extrapolate from a single screen shot to a release schedule), VMware folks will stop saying "Microsoft is copying what we do" and start saying "Microsoft doesn't do all the things we do" and the argument will move to how good or bad paging in the Virtualization layer is really (VMware do it, Microsoft don't).

]]>