I fully admit that I'm lacking in virtual machine knowledge, but some of those "features" mentioned seem to be useless. Hot-add CPU and memory? Where will that be useful? Let me rephrase the question. What operating systems will be able to use that feature? How is Windows going to act when one nanosecond it's operating on a 2-CPU system with 4GB of memory, and the next nanosecond it's operating on a 16-CPU system with 32GB of memory?

64 cores is a record?

Umm, let's see. MS are claiming a record for a product that hasn't shipped yet.

Are there any off the shelf 64-way Intel or AMD servers out there yet? Or are MS "accidentally" counting cores instead?

More vapourware I suppose.

IBM have been shipping their p595 for a while now. This goes up to 64 CPUs. So at best M$ can only be claiming they are equalling a record.

"How is Windows going to act"?

I am not a virtual machine expert either, but I believe you are simplifying somehow, leaving out some of the VM advantages.

But anyway - to answer you naive question - haven't you ever seen BSOD ?!?! :-)

How will Windows react when running on 32 Gb and 16 cpu's and a nanosecond later gets downgraded to 4 Gb and 2 CPU's?

I would guess that the hotplug process would probably be more of a warmplug where the system is informed before adding/removing hardware as to be able to copy data in RAM elsewhere, or to be able to move execution around to different cpu's - This could either be hardware based (some microswitched cover that needs to be moved before the user is able to access a hotplug ram module or cpu, otherwise, think USB : Right click on the task bar, select ram chip/cpu to remove, and Windows will tell you after a few seconds that it is safe to remove that module from the system (just don't get it wrong, stop chip 1 then remove chip 2, as then you will get a virtualised blue screen :)

Cheers,

Daniel

The use is easy; You use RAID for your hard drives but when a RAM chip goes bad or a CPU cooks, the system stops.

In this sort of configuration, the OS could detect a failure condition (a sort of S.M.A.R.T for RAM and processors) and disactivate the component and warn the administrator. The load would be shared over the remaining ressources until the faulty module can be replaced, on the fly.

This would be very useful for any high availability setup or for remote offices with no onsite support, as it would allow you to continue operations and give you time to grab spares and a plane ticket and not have a) the office/website/lab/nuclear detection system shut down, and b) having to explain to the most IT competent guy onsite how many beeps the box makes when they plug it in, and not having to manage a multi-machine failover cluster would also be a bonus (think of it as a cluster-in-a-box)...

Cheers,

Daniel

Crash

....and ITs Stations....Virtually Controlled Hubs.

"Those following this market closely know that the virtualization wars have only just begun."

Games, Gentlemen, not wars.

Now who would relate that Virtualisation is not Realised.

Good Morning, MicroSoft. Do you want to Play the God Macro? AI Campbell does IT. Or is IT Rupert @ Work, Rest and Play Station 2?

Daniel,

Spoken like a native. I concur.

For some ideas on how MS may end up handling some of the issues, here is how IBM does it.

On the RS6K (Pseries) side of things, the hypervisor is an important key part and the OS needs to be capable of handling the changes the hypervisor makes. I think on the x86 side this will end up being the case as well.

Partitions set up with the AIX operating system support dynamic allocation of RAM (add, subtract, move). Last I saw Linux on POWER could not handle dynamic allocation of RAM.

Resources (adapter or processor) can be added, moved or removed dynamically to both AIX and Linux. The target device (for adapters) needs to be removed from the running configuration inside the OS first before you can tell the hypervisor to move the device.

Processors can be added or subtracted with some loss of performance (you now either lost some cached data or added an empty cache to fill) and extra context switching depending on what you do.

This is all at the dedicated level. When you get into logical partitions and virtual devices it gets even more fun, but all the same stuff can be done.

Dynamic reallocation of resources is great because you do not need to bounce the running system to change things.

So Windows will have to support the ability to add/remove/move memory/processors/devices (even when virtualized) and there will need to be a decent hypervisor behind it all

As for memory fault tolerance

On it’s Xseries (x86) hardware they like to use chipkill RAM. It will be able to handle failure of a single memory module and keep on trucking. Each mem chip has an extra 8 bit chip and it acts as sort of a parity/raid in RAM.

As someone who's deployed systems which support thousands of VMs on VMWare ESX, I can shed some light. Hot-add means exactly that. ADD, not subtract. Adding new hardware resources, whether real or virtual, is much easier than subtracting. This feature is for management of a loaded system where the workload is greater than anticipated in a specific VM, so that you can allocate extra RAM and CPUs. You can also grow your VM hosts in the same way. Windows was the first x86 server OS to support hot-add RAM, and apparently now hot add CPU. Because of the very basic virtual hardware presented to the VMWare VMs, they cannot support hot add RAM without a software upgrade to their virtual hardware.

And FWIW, VMWare ESX performed better than Virtual Server, but the Linux based software showed us the PURPLE Screen of Death many times, with nary a BSOD from the Win2K3 guests or Virtual Server hosts. Vmotion and Virtual Center were just too good to pass up.

"Microsoft should be allowed a large amount of leeway in this market where it's the true underdog"

What??? So regardless of who the company is or how good / late their product / technology is, they shoud be allowed leeway becasue they're the underdog?

I thought you guys were more sophisticated than that. [disappointed]

And anyway, a company that can crush a technically superior competitor with its fiancial and marketing might probably shouldn't be considered an underdog...