The ROI will have to be compelling and the risk to disruption minimal. At this point I’ve not heard anyone articulate how exactly this will happen.
In addition, one has to question the logistics of how competing ARM vendors will work together in a mixed customer environment, given they all have their own unique implementations and fabric strategies.
“If you look at the number of companies looking to do microservers, the market starts looking crowded in a hurry,” said analyst David Kanter, noting that at best there were 30 potential customers for Microservers, though he estimated the real number may be closer to 15.
Kanter explained that to have any hope of success in the market, companies who wanted in would need a serious breadth and quality of IP for CPUs, caches, memory, interconnects, networking, etc.
Indeed, in 2014, servers are going to need large L3 caches, DDR4 memory controllers, 10GBE or Infiniband networking, coherent interconnets (e.g., QPI), PCI-E gen 3, not to mention comprehensive power management and RAS features.
“Some of those pieces you can license, but quite a few you cannot, and even licensing something may not be quite as good as doing a custom design,” said Kanter, noting that most ARM licensees fell short in this respect, and that only AMD and Applied Micro really had what it took at this point in time.
“Applied Micro is interesting because it is using its own custom CPU core that is higher performance than ARM's reference designs,” he said, noting that this gave the firm an advantage in terms of performance and cost structure, should they achieve sufficient volume.
This week, we’re going to see yet another media cycle around microservers as Intel prepares for the upcoming launch of Centerton, the firm’s first server class SoC for the data center. This will be followed by the company’s 22-nm Avoton product in 2013, which will almost assuredly have a follow-on in 2014.
That’s a grand total of three Intel product cycles before we even see our first real ARM server product enter the market.
While nobody questions that Intel most certainly has its work cut out for it in the ARM dominated world of smartphones, the firm has proven it can deliver a competitive product, albeit not at scale yet. Now the pressure is on Intel to build market share and customers in the coming years.
The path forward is equally daunting in the data center and arguably less forgiving for ARM. Enterprise IT managers demand a robust TCO model, have little patience for downtime and ARM is unproven in these areas of real-world deployments.
It may also not be worth the effort.
“If you believe that ARM-based servers can grab 10% market share, how much *silicon* revenue is that? And how many SoC companies can it sustain?” asked Kanter, pointing out that in a market of only about 10 million units, there wasn’t room for five to eight companies.
“If you look at it that way, it's starting to sound like a pretty low volume business,” he said.
The ARM eco-system is coming to the data center, that much is certain, but the road traveled to get there will be a long and challenging one with more competitive pressure than the firm has ever felt before. And success may not taste as sweet as the firm thinks it will be.
Solely considering Intel Xeon production, the total Sandy Bridge E Series market for dense ARMs on Blade is 113,594,361 units valued at $72,395,536,747 revenue potential.
Xeon 12xx/24xx/26xx production volume priced less than $250 = 7,466,263 units.
Xeon 12xx/24xx/26xx/46xx production volume priced less than $625 = 24,451,308 units.
ARM Ltd. has apparently been under financial pressure lately. And got caught drinking its own Koll aid in two of four attempts to hit high volume royalty home runs out of the ball park.
Of these only one brought in base holders, one struck out on projections, one made first base, and for the final batter up . . .
Noteworthy while ARM Ltd. was out aggrandizing that market communications could resolve what only world class platform development can secure, beyond smart phone & tablet, neglected its entire business foundation as Intel stormed every industrial beach head which were left not only unguarded by ARM executive management; but unattended.
In setting commercial market expectations so high ARM Ltd. forced a financial cost onto some of its customers seeing their passing. And for reasons that solely have to do ARM Ltd. believing they could talk the market into accelerated Intel victories lacking any market foundation what so ever.
Press relations cannot make a product, or a successful product market. Innovations take time, nurturing the market potentials and not waking up Goliath so early.
ARM Ltd. has been caught in the trap of its own public relations puffery with nothing to show for it.
This is surely an embarrassment for an enterprise capable of world class management with oversight supporting compliment development while guarding against the competitive financial risks.
The executive management of ARM Ltd. will never make this mistake ever again.
AUTHOR:" But lets face it, Microservers is not even that big of a market anyway, with only a handful of important customers like Facebook, Google and Amazon. " Doesn't Google have the worlds largest data center already? Or close to it. Microservers WILL BE the only market next decade that has any growth in "server" segment in my opinion. And the sky is the limit for clouds (pun intended).
Of course it remains to be seen whether ARM will deliver on promises, but they did well so far. ARM claims pretty good performance for their 'microserver' chips: 1250 SpecInt2000 for 1.7GHz chip. It is weird that they would quote SpecInt2000 because it's been superseeded by SI2006 six years ago, so maybe those numbers should be taken with some caution, but if you take them on face value, they are pretty good: Intel 1.7GHz chips of that era (e.g. P4) tend to present at about half the performance.
Even after allowing some 'benchmark shrinkage' and improvements in Intel architecture since that time, ARM should be competitive with Intel on performance. Based on their track record, they should do fine on power. As for the price, Intel quoted unit price for their new SoC announcements was above $50---again, ARM should be quite competitive, especially considering that ARM chips are traditionally pretty integrated, with the equivalent of North/South bridge on chip, whereas Intel usually requires separate motherboard/system chipsets.
Having said that, I am reminded of the story I heard from one of the RISC pioneers. He showed up at a Vax shop and demonstrated his machine. Afterwards, the customer said to him: 'So, your machine is twice the speed at half the price of what I have now. Give me one good reason why I should buy it".
Hey Whorton's Horta1212, Check my comment above. Cortex-A15 at 1.8GHz consumes 6W. Its a Fact. When Intel has a SoC for 6W, tell ARM to jump off the cliff. What are you smoking with your x86 power hungriness. Everyone knows ARM can not design a performance efficient design. You are duped by the ARM smoke screen
It's pretty simple really. It's collection of chips that have many cores that run very efficiently. It would be like taking today's (or yesterday's) laptop chips, stripping down many peripherals, creating an interconnect and sticking it within a single chip. The major advantage of it should be much better performance/watt than current server chips, which have been designed for years to be speedy, power-hungry beasts. What allows ARM microservers to achieve these goals should be better process technology and the ability to sidestep x86's power hungry ways (by not being x86).
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.