As cloud computing-fueled devices like the iPad grow in popularity, so will associated greenhouse gas emissions, according to Greenpeace's "Make IT Green" report. The report, which dubs 2010 the Year of the Cloud, offers up a disturbing statistic: Cloud computing greenhouse gas emissions will triple by 2020.
The increase in emissions makes sense. As we increasingly rely on the cloud to store our movies, music, and documents, cloud providers will continue to build more data centers--many of which are powered by coal. Facebook, for example, recently announced that is building a data center in Oregon that will be powered mostly by coal-fired power stations, much to the chagrin of groups like Greenpeace.
The solution to the cloud computing problem is fairly obvious. Greenpeace explains in its report, "Companies like Facebook, Google, and other large players in the cloud computing market must advocate for policy change at the local, national, and international levels to ensure that, as their appetite for energy increases, so does the supply of renewable energy." As we've noted before, companies like IBM, Google, and HP have already begun to make strides in cutting data center energy use. But there is still plenty of work to be done--as it stands, the cloud will use 1,963.74 billion kilowatt hours of electricity by 2020.
Intel's switch to the Nehalem architecture was finally completed Tuesday with the launch of the Nehalem-EX Xeon 6500 and 7500 processors, the last of the Core, Xeon, and Itanium chips to get the Quick Path Interconnect and a slew of features that make Intel chips compete head-to-head with alternatives from Advanced Micro Devices. The price war at the midrange and high-end of the x64 market can now get underway, while the all-out, total price war awaits the debut of AMD's Opteron 6100 processors in the second quarter.
Since the summer of 2008. Intel has been previewing its top-end, eight-core Nehalem-EX beast, which we now know as the Xeon X7560. As it has done with prior generations of Xeons, the Nehalem-EX line is not comprised of one or two chips, but a mix of chips with different features (clock speed, cache memory, HyperThreading, and Turbo Boost) dialed up and down to give customers chips tuned for specific workloads.
While last year's Nehalem-EP Xeon 5500 and this year's Westmere-EP Xeon 5600 processors are aimed at workstations or servers with two sockets, with the Nehalem-EX lineup, Intel has broadened the definition of its Expandable Server (this is apparently what EX is short for, with EP is supposed to be an abbreviation for Efficient Performance) to include two-socket machines as well as the four-socket and larger machines that prior generations of Xeon MP processors were designed for.
Intel, no doubt, would have preferred to keep the Xeon DP and Xeon MP product lines more distinct, and charged a hefty premium for machines that needed expanded processor sockets or memory capability. But server makers and their customers were having none of that. With the rapid adoption of server virtualization and the need for larger memory footprints even for two-socket boxes, the Nehalem-EX processors have been tweaked so they can be used to support very fat memory configurations on even two-socket workhorse servers. This will eat into the volume Xeon 5500 and 5600 market, to be sure, but it is better to sell a Xeon 6500 or 7500 server in a two-socket box than have a customer dump Intel for AMD.
The Xeon 6500 and 7500 processors will also blur some lines between Xeon processors and the former "flagship" Itanium processors, which were supposed to take over the desktop and server arena starting a decade ago, but have been relegated mostly to high-end servers from HP running HP-UX, NonStop, and OpenVMS at this point in their history. The Itaniums were distinct in many ways from the Xeons, but the main distinction they held was better reliability, availability, and serviceability (RAS) features than Xeons had, and on par with mainframe, RISC, and other proprietary architectures from days done by.
The eight-core Nehalem-EX Xeon 7500 beast
But at the launch event today in San Francisco, Kirk Skaugen, vice president of the Intel Architecture Group and general manager of its Data Center Group, made no bones about the fact that the Nehalem-EX processors and their related Boxboro chipset that is shared with the Itanium 9300 processors launched in early February have common RAS features.
The new chip, explained Skaugen, has 20 new RAS features, including extended page tables and virtual I/O capabilities as well as a function that is in mainframes, RISC iron, and Itaniums called machine check architecture recovery, which allows a server to have a double-bit error in main memory and cope with it without halting the system. With Windows, Solaris, and Linux supporting these RAS features, as well as VMware's ESX Server hypervisor, this makes servers based on the Xeon 7500s just as suitable a replacement for proprietary midrange and mainframe platforms and RISC/Unix servers as the formerly beloved Itaniums.
Skaugen said that the Nehalem-EX chips would allow server makers to create two-socket servers that support up to 512GB of main memory, nearly three times as much as AMD can do using 8GB DIMMs with the Magny-Cours Opteron 6100s announced yesterday. Intel will be able to support 1TB of main memory in a four-socket configuration, while the controller inside the Opteron 6100 only allows a four-socket machine using these chips to address a maximum of 512GB.
Skaugen rubbed it in a little that Intel's Nehalem-EX partners had over 50 new products in rack, tower, and blade form factors, and that it had 75 per cent more four-socket designs than with any prior server chip launch in its history. A dozen OEM partners have 15 different servers in the works that will span eight or more processor sockets, and apparently some are pushing their designs up to 16, 32, or 64 sockets.
The big bad box at the Nehalem-EX launch, of course, was the Altix UV massively parallel supercomputer, which El Reg told you all about last November. The Altix UV machines allow for up to 2,048 cores (that's 256 sockets and 128 two-socket blades) to be lashed together in a shared memory system suitable for running HPC codes. The shared global memory is not the same as a more tightly coupled symmetric multiprocessing (SMP) or non-uniform memory access (NUMA) cluster used in general purpose servers for running applications and databases. But that said, the Altix UVs are very powerful machines indeed and are intended to scale to petaflops of performance.
The Boxboro chipset that Intel is shipping as a companion to the Nehalem-EX chips supports configurations with two, four, or eight sockets gluelessly. If you want more sockets than that, you have to create your own chipsets, as HP, IBM, Silicon Graphics, and Bull have done for sure and others will no doubt follow. But you can't just plug any old Nehalem-EX chip into any old configuration. That would be too simple, and Intel likes to charge premiums for features, like most capitalists. Take a gander at the feeds and speeds of the Nehalem-EX lineup:
The Intel Nehalem-EX Xeon 7500 and 6500 processors
The first thing you will notice is that there are two different families of Nehalem-EX processors. The Xeon 7500s are aimed at general-purpose workloads and offer the most socket expandability. All of these chips can be used in two-socket or four-socket boxes, and some of them can be used in eight-socket or larger machines, too. The Xeon 6500s are cut-down versions of the chips that only work in two-socket boxes and that are specially tuned for the HPC market. These chips, explained Skaugen, were optimized to have the highest bytes per floating point operation ratio while minimizing the amount of node-to-node communication among the processors in the complex.
The top-end X7560 part has eight-cores spinning at 2.26GHz, has 24MB of L3 cache on the chip, and is rated at 130 watts using Intel's thermal design point (TDP) scale. The chip supports Turbo Boost, which allows for a core to have it cycle time jacked up if other cores are shut down when they're not being used, and it also supports Intel's HyperThreading simultaneous multithreading, which virtualizes the physical pipeline in the chip so it looks like two virtual pipelines to a system's operating system and its applications. In best-case scenarios, HT can boost performance of applications by around 30 per cent. In 1,000-unit trays, the per-chip price for the X7560 is a whopping $3,692. That is exactly what Intel charged for a dual-core Montvale Itanium 2 with 24MB of L3 cache.
The X7550 drops the clocks down to 2GHz, chops the L3 cache down to 18MB, and the price comes down to $2,729, which is exactly what Intel was charging for its top-bin six-core Dunnington Xeon X7460 processor running at 2.66GHz with 16MB of L3 cache. The next part down, the X7542, jacks the clocks up to 2.66GHz, drops the cache down to 18MB, cuts out HyperThreading, and reduces the core count down to six from eight; the price drops down to $1,980.
For that same $1,980 you can get a standard 105 watt part, the E7540, running at 2GHz with six cores and that same 18MB cache. If you are willing to take lower clock speeds, you can get even cheaper standard parts, the E7530 and E7520, which cost $1,391 and $856, respectively. Intel has also cooked up two low-voltage parts, the L7555 and L7545, running at 1.86GHz and rated at 95 watts, which have eight and six cores, respectively. These are reasonably pricey chips that will no doubt be used inside Nehalem-EX blade servers where a premium is expected in exchange for extra density.
Generally speaking, the Xeon 6500 processors are cheaper than their Xeon 7500 counterparts because they have some features and functions turned off, as El Regpredicted they would last fall. This is in keeping with the general philosophy that HPC shops are super-stingy and will not pay one extra penny for a feature they don't want and will never use.
The Nehalem-EX processors are implemented in 45 nanometer processes and have 2.3 billion transistors. ®
When Steve Jobs met Google boss Eric Schmidt for coffee late last week, they may or may not have reached some common ground on certain hot-button subjects. But odds are, they didn't see eye-on-eye on Adobe Flash. As Jobs prepares to ship his much ballyhooed Apple iPad without even the possibility of running Flash - which he calls "buggy," littered with security holes, and a "CPU hog" - Google is actually integrating the beleaguered plug-in with its Chrome browser.
With a blog post on Tuesday, Mountain View announced that Flash has been integrated with Chrome's developer build and that it plans to offer similar integration with its shipping browser as quickly as possible.
Google has been known to say that HTML5 is the way forward for internet applications. But clearly, it believes in the plug-in as well, and it has no intention of pushing all development into the browser proper. "Just when we thought that Google was the champion of HTML5 they turn around and partner with Adobe on Flash to ensure that the web remains a mess of proprietary brain damage," one netizen said in response to Google's post.
Last summer, Google proposed a new browser plug-in API, and with today's blog post, it also said that Adobe and Mozilla have joined this effort. "Improving the traditional browser plug-in model will make it possible for plug-ins to be just as fast, stable, and secure as the browser’s HTML and JavaScript engines," the company said. "Over time this will enable HTML, Flash, and other plug-ins to be used together more seamlessly in rendering and scripting.
"These improvements will encourage innovation in both the HTML and plug-in landscapes, improving the web experience for users and developers alike."
What's more, Mountain View is developing a native code browser platform of its own, dubbed Native Client. This is already rolled into Chrome, and it will be an "important part" of the company's browser-based Chrome operating system, set for launch in the fall.
By integrating Flash with Chrome, Google said that it will ensure users always receive the lastest version of the plug-in and that it will automatically update the plug-in as needed via Chrome's existing update mechanism. And in the future, the company added, it will include Flash content in Chrome's "sandbox," which restricts the system privileges of Chrome's rendering engine in an effort to ward off attacks.
In July, with a post to the Mozilla wiki, Google proposed an update to the Netscape Plug-in Application Programming Interface (NPAPI), the API still in use with browsers like Chrome and Firefox, and both Adobe and Mozilla are now working to help define the update.
"The traditional browser plug-in model has enabled tremendous innovation on the web, but it also presents challenges for both plug-ins and browsers. The browser plug-in interface is loosely specified, limited in capability and varies across browsers and operating systems. This can lead to incompatibilities, reduction in performance and some security headaches," Google said today.
"This new API aims to address the shortcomings of the current browser plug-in model." The new setup was developed in part to make it easier for developers to use NPAPI in tandem with Native Client. "This will allow pages to use Native Client modules for a number of the purposes that browser plugins are currently used for, while significantly increasing their safety," Google said when the new API was first announced.
Native Client and NPAPI have been brewing for months upon months, but today's Chrome announcement would seem to be a conscious answer to Steve Jobs' hard-and-fast stance on Flash. Presumably, the company sees this a way to ingratiate existing Flash shops who've been shunned by the Apple cult leader. One of the many questions that remain is whether Chrome will give users the option of not installing Flash. With the new developer build - available here - you must enable integrated Flash with a command line flag. ®
The eight-core Nehalem-EX Xeon 7500 beast
The Intel Nehalem-EX Xeon 7500 and 6500 processors