It looks like server virtualization really is heading for the mainstream on Power Systems machines. If some figures provided by the top brass at IBM are any indication. While logical partitioning has been available on OS/400-based servers since V4R4 was launched with the Northstar PowerPC servers back in 1999, virtualization on …
Timmy on big blue payroll ?
No ? So Timmy is just a big blue hugger, is it.. ?
A really, really boring product placement article.. at times I wondered if I am watching James Bond peddling Aston Martin - without the babes or the action, of course.
Customers buying != customers using... I'm a cynical type I know, but to me this could be either:
i) IBM Power sales are incetivised to sell PowerVM - and tuck it into customer deals whether they need ot or not.
ii) THe configurator that IBM uses for producing Power system quotes automatically adds PowerVM to every quote. As most quotes go to dumb purchasing teams now ratherv than IT folks, this never gets picked up.
Or maybe customers really are using more? But sales stats aren't a sensible way of measuring this.
IBM Shops swallow it hook line and sinker
seriously, virtualizing on a proprietary platform like p-series or i-series is a bit like going to a 5 star restaurant but just ordering a salad to save money.
DEC/Compaq did it on the Alpha before IBM
The OpenVMS/Tru64 crowd was able to do this even earlier.
SUN has been shipping systems a free Hypervisor for years...
I find it interesting that SUN & Fujitsu have been shipping a free hypervisor with it 1-4 socket CoolThreads processors for years.
IBM continues to "get with the program"!
Well we are using it, everywhere we can. I mean we are replacing p690's with power 560/ 550'es where just savings in maintenance is paying for the damn migration project. And yes the POWERVM is a new marketing name but the basic stuff have been on servers since 2001, and the main functionality of the current powervm have been around since 2004.
Well, today VMS doesn't even support vpars on Integrity servers afair, which is kind of a bugger. And watch out a Mainframe guy might hear you and start telling about how they virtualized back in the 60'es before even I was born.
The style hypervisor that is shipped with Niagara based boxes is a logical partitioning technology hence you divide the machine into smaller bits, allocating CPU threads to different domains. That type of hypervisor have been shipping on POWER servers since 2001. Now the hypervisor that started shipping with the POWER5 boxes in.. 2004 I think it was, is what is using what Big Fat Blue is calling shared pool logical partitioning (SPLPAR), here you just allocate shares of CPU ressources, and when a partition isn't using them others can. The same functionality that you would implement using containers on a solaris system. In my book virtualization on power is years in front of the, just as clear number 2, which is SUN and HP really have to get things going with regards to virtualization.
I don't really think people realize that by embracing virtualization like people that build POWER and also Solaris solutions have done, is actually hurting IBM and SUN hardware sales numbers.
Gartner and IDC have called this effect the virtualization effect, and it's also hurting the x86 sales bigtime.
And it really isn't that strange, cause you are pumping up the utilization of the physical hardware when you are using virtualization. So if you had a workload that ran on a 24 core system using partitions and with a 20% average utilization. If you now can get a system with cores of twice the speed and you can use virtualization to raise the average utilization to 60% by letting the partitions that don't use 'their' CPU ressources, let other partitions use it.
Then you are actually reducing your needs for compute power from 24/(2x3)=4cores, which means that you can jump to a p550, if we use POWER as an example.. Now if you had just made a 1-1 migration (lpar->lpar) to a new box with cores that were twice as fast you would have needed 12 cores. Then you would most likely have bought a power 570. And the price of a 4 core power550 is something completely different to a 12 core power 570.
@Jesper Frimann--- hypervisor question
David said, "SUN & Fujitsu have been shipping a free hypervisor with it 1-4 socket CoolThreads processors for years."
Jesper said, "The style hypervisor that is shipped with Niagara based boxes is a logical partitioning technology hence you divide the machine into smaller bits, allocating CPU threads to different domains. That type of hypervisor have been shipping on POWER servers since 2001."
Has IBM shipped it for free, or has IBM been charging people for it since 2001?
(I was under the impression that they were charging for it... I could be wrong, and if so, I stand duly corrected!)
Jesper said, "I don't really think people realize that by embracing virtualization like people that build POWER and also Solaris solutions have done, is actually hurting IBM and SUN hardware sales numbers."
SUN predicted that virtualization was going to hurt their hardware sales, but they also thought it would drive their hardware sales, in the low end. I personally think virtualization hurt Sun's sales more than they expected.
Most of my applications run with thousands of threads and hundreds of processes, so they scale very well with CoolThreads... actually, better than I expected. I designed one of my solutions to leverage Solaris 10 Containers... it also runs better than I expected. The scalability of the V890 had really enabled me to skip upgrading to the M series and hold out for something like the 4 T2+ CPU platforms or something later.
If RCK shows up in 2009, great, if not, the 4xT2+ will be enough for my applications, running LDom's and Containers.
Prices on powervm
"Has IBM shipped it for free, or has IBM been charging people for it since 2001?"
Brrr.. that was a good question. I cannot say for sure. But I have looked at some of the old configurator output I got from IBM Sales for solutions I've done in the past.
And on all the POWER4 solutions p690'es there is no virtualization charge, and for the power5 solutions I have on the HD there isn't either.
Now for the power6 there are charges after POVERVM became a products, but on blades you get a standard edition of POVERVM for free.
As far as I know and can see from the SUN and IBM homepages, then the features that are a part of the hypervisor on the T5XXX series are also free on a POWER servers. You can do logical partitioning virtual networks. So the free part on power servers are at least equal to the free part on T5XX servers. Now wpars, which is akin to containers are also included for free in the OS.
But when you want shared pool lpar, Virtual IO servers, partition migration wpar dynamic movement of wpars etc. then on most of the servers you will have to put money on the table.
It looks to me, from my understanding of things, that IBM have made sure that what you get for free on T5XXX'es you also get for free on power boxes. And then the fancy stuff you have to pay for, not really surprising. And I have no problem paying for Shared Pool Logical Partitioning, it pays for itself many times over.
"SUN predicted that virtualization was going to hurt their hardware sales, but they also thought it would drive their hardware sales, in the low end. I personally think virtualization hurt Sun's sales more than they expected."
I must admit I have trouble seeing where SUN will make the big bucks on their current strategy.
"If RCK shows up in 2009, great, if not, the 4xT2+ will be enough for my applications, running LDom's and Containers."
'good enough' is the key word here. You have to remember that the other hardware vendors also move forward. If Rock comes out in late 2009, early 2010 it will be facing POVER7. Which looks a bit like a monster.
RCK, Itanic, and POWER Speculation
Dave commented, "If RCK shows up in 2009, great, if not, the 4xT2+ will be enough for my applications, running LDom's and Containers."
Jesper commented, "'good enough' is the key word here. You have to remember that the other hardware vendors also move forward. If Rock comes out in late 2009, early 2010 it will be facing POVER7. Which looks a bit like a monster."
I agree. POWER looks very competitive on the horizon.
Sun has been spot-on with every release of the CoolThreads processors, releasing early instead of late.
A 16 core SPARC CoolThreads, a 16 core SPARC RCK , a Fujitsu VII+, a POWER 7, and 4 core Itanium may all show up this year or early next - I don't count any of them out.
If course, AMD and Intel are fighting it out for x64 dominance. I am hopeful for AMD, since they singlehandedly kept the x8 architecture from going the way of the dinosaur (by releasing a 64 bit architecture and forcing Intel to build 64 bit extensions.)
Any of the vendors could have a silicon failure, causing severe problems with their road maps. Sun and Fujitsu are most able to deal with a silicon failure, with SPARC T, SPARC RCK, and SPARC64 VII+ in their future. SPARC vendors are triple-sourced, with some level of overlap.
Hybrid x64 manufacturers like IBM, HP, and Dell are well positioned to deal with silicon failures, since they are dual-sourced.
Single sourced vendors like IBM POWER and HP/Intel Itanium are least able to deal with a silicon failure. IBM and Intel have deep pockets, which are not so deep during global recessions like we are having now. A silicon failure could be catastrophic for AIX or HP-UX during times like this.
As the old saying goes, "it's not over, until the fat lady sings!"
"I agree. POWER looks very competitive on the horizon."
Lets correct that to POWER is very competitive.
"Sun has been spot-on with every release of the CoolThreads processors, releasing early instead of late."
Jup as far as I know this is correct. but that hasn't been the case with other SUN processors, for example Rock.
"Any of the vendors could have a silicon failure, causing...."
Now that is what I call speculation. You could say that HP have suffered from setbacks on Tukwila, it's still not here but their UNIX sales numbers are decent. They havent dropped as much as SUN's have. So you can still have a success full UNIX business without really renewing you products.
With regards to POWER I would say that IBM have quite a few nobs to turn on power6 that could extend it's life. And then cover up a failure of power7. First of all they could enable all memory controllers and all the memory channels on each chip. Or they could simply make a QCM version like they did with POWER5+. The 1.8Ghz p560q is still faster than the POWER6 power 560 on a per socket basis. U need a 5.0Ghz power 570 to get better per socket performance.
So for example 4-6 core QCM/HCM module running at 3 GHz would still be pretty competitive. And I must admit I simply don't understand why IBM haven't made a QCM version of POWER6.
@Jesper Frimann--- Well ; RCK, Itanic, and POWER Speculation
Dave said, "Any of the vendors could have a silicon failure, causing...."
Jesper said, "Now that is what I call speculation."
LOL! That's why I titled the post as "Speculation!"
Jesper said, "You could say that HP have suffered from setbacks on Tukwila"
I could have also mentioned the AMD Quad-Core setback, as well.
Jesper said, "but that hasn't been the case with other SUN processors, for example Rock"
That possible silicon failure was in-scope when I made that broad sweeping statement.
Intel gave room to AMD, IBM, and SUN to catch up once they canceled their faster speed processors to re-engineer for multi-core, since they had all been moving towards 64 bit multi-core while x86 was stuck at 32 bit single core. While this thought was in my mind, I did not have a way to tie that into a short but broad statement.
The speculative statement really applied to multiple vendors, but I didn't want a backwards looking accepted piece of history to dominate my forward looking considerations.
Jesper said, "I must admit I simply don't understand why IBM haven't made a QCM version of POWER6."
I am sure it is related to where they are investing and how it fits into the line... no vendor wants to cannibalize coming high end profits with a lower end architecture. I suspect their business model has been profitable.
- Comment Renewable energy 'simply WON'T WORK': Top Google engineers
- Useless 'computer engineer' Barbie FIRED in three-way fsck row
- Game Theory Dragon Age Inquisition: Our chief weapons are...
- 'How a censorious and moralistic blogger ruined my evening'
- Amazon warming up 'cheapo web video' cannon to SINK Netflix