September 19, 2011
I just attended IDF and I’ve got to say, Intel has certainly gotten the cloud message. Almost everything is centered on clouds, from the high-concept keynotes to the presentations on low-level infrastructure, although if you dug deep enough there was content for general old-fashioned data center and I&O professionals. Some highlights:
Chips and processors and low-level hardware
Intel is, after all, a semiconductor foundry, and despite their expertise in design, their true core competitive advantage is their foundry operations – even their competitors grudgingly acknowledge that they can manufacture semiconductors better than anyone else on the planet. As a consequence, showing off new designs and processes is always front and center at IDF, and this year was no exception. Last year it was Sandy Bridge, the 22nm shrink of the 32nm Westmere (although Sandy Bridge also incorporated some significant design improvements). This year it was Ivy Bridge, the 22nm “tick” of the Intel “tick-tock” design cycle. Ivy Bridge is the new 22nm architecture and seems to have inherited Intel’s recent focus on power efficiency, with major improvements beyond the already solid advantages of their 22nm process, including deeper P-States and the ability to actually shut down parts of the chip when it is idle. While they did not discuss the server variants in any detail, the desktop versions will get an entirely new integrated graphics processor which they are obviously hoping will blunt AMD’s resurgence in client systems. On the server side, if I were to guess, I would guess more cores and larger caches, along with increased support for virtualization of I/O beyond what they currently have.
One of the highlights of Paul Otellini’s keynote was a demonstration of a solar powered CPU, powered by a small solar cell and a lamp, that ran Windows. Some of the press has incorrectly reported this as a variant of the Sandy Bridge – it was actually a pure research implementation of an older Pentium III design with less than 1/10 the number of transistors as the newer CPUs. It was intended as proof that Intel can design very low power systems and backup for Mr. Otellini’s claim that 2012 would be the year of the tablet and the mobile phone for Intel. They certainly will be in a better place with the newer products, but I am still skeptical.
SSDs and NVRAM were a relatively high-profile topic, with sessions on the economics and futures of SSDs. To net it out, SSDs have years to run before they run into any technology walls, and when they do (or possibly in parallel), there are new NVRAM technologies, both conventional and things like memrister and phase change technology, to keep the pot boiling.
While all of their major systems partners brought servers to the partner showcase, the big emphasis this year was on client systems particularly the new UltraBooks, systems developed by multiple vendors from Intel’s Windows-derivative reference design strongly influenced by the MacBook Air. UltraBooks look like the MacBook Air, with thin, often wedge-shaped profiles and weighing under 3 pounds for a 13” screen, and have the new Core i3 and i5 CPUs, 4 – 8 GB of RAM, and up to 128 GB of SSD. Intel and Microsoft have done some magic with Windows, and these systems will suspend when the lid is closed and resume with no loss of state within 3 to 5 seconds after you re-open the lid. Very cool, will hit the market (Samsung has theirs, the Series 9, already available). On my Xmas list this year for sure, and my opinion is that they will blunt some of the momentum of tablets – I was planning to buy the next generation iPad, but a look at the UltraBooks may have changed my mind.
On the server side, all of the major systems vendors had dense server systems sporting the cloud label, and both Citrix and VMware along with hardware partners were touting both VDI and cloud management software. Security was a high-profile topic in many of the partner exhibits as well as Paul Otellini’s keynote.
Data center and I&O stuff
Some interesting sessions on data center management, particularly power and cooling technology management, with Intel offering itself as an example. I found the session on Intel’s own internal best practices particularly compelling since their data center strategy leaves them with approximately 90 widely distributed data centers of modest size, and thus more representative of best practices for many mainstream I&O groups than do some of the profiles of Internet mega data centers. Intel has developed a body of practice around designing for very dense racks, in this case up to 25 KW per rack and using conventional air cooling rather than exotic add-on heat exchangers. They presented a body of best practices and case studies of successively increasing density, and indicated that they are confident that they can cool standard IT (mostly servers) at up to 35 KW per rack with these guidelines. Worth checking out, including a book by the presenters, “Creating the Infrastructure for Cloud Computing.” And not to worry, it is really about data centers in general, not specifically cloud, but I guess you gotta go with the flow on titles.
As always, IDF was a tour de force of Intel technology, replete with impressive demonstrations of advanced technology and an underlying drumbeat that the Intel juggernaut will continue to crush any opposition. In general, not a lot to argue with on the general thesis that they are the top dog in manufacturing, have an aggesssive road map, and high aspirations to capture an immense number of design wins as the number of connected and sensor-enabled and mobile devices continues to mushroom.
On the other hand, I think they may face more competition than they expect from AMD in HPC and cloud as opposed to general enterprise IT (a claim that is at least partially substantiated by the latest set of Top 500 results, which shows AMD gaining some relative traction in this rarified upper tier of the HPC worlds), and ARM, MIPS and others will continue to make their life miserable in the mobile device market, particularly since Microsoft has signaled very clearly that Windows 8 will not be an x86-only proposition.
My net take is that Intel will prosper, and the silly media hounds that are predicitng its partial collapse are barking up the wrong tree – expect them to continue as the dominant player in a tough competitive environment.