Intel’s New Shopping Craze: Wireless Technologies For Smartphones

August 5, 2010

Remember the late 90′s, when Intel was throwing billions in its quest to become a telecommunication powerhouse, to finally write it all off?

Well, sounds that the chipmaker is at it again with the recent acquisitions of Infineon’s cellular and GPS technology as well as the assets of small 4G vendor Comsys.

Hopefully, the mastermind of Intel’s first telecom foray, Sean Maloney, might remember one or two things learned from his first failed attempt to build a communications empire!

As analyst Linley Gwennap writes it,

“Making it clear that money is no object in its quest to become a major player in the smartphone market, Intel plans to acquire the wireless operations of Infineon in a deal valued at more than $1.2 billion. Infineon’s cellular and GPS technology will complement Intel’s Atom processor and Wi-Fi expertise, allowing Intel to deliver a complete solution for the rapidly growing smartphone and tablet-computer markets.”

Infineon ranked fourth in cellular-baseband shipments last year with 10.7% unit share, according to a recent report from The Linley Group. Infineon is a major supplier to Apple, Nokia, and Samsung.

But “the German vendor lacks an application processor, however, making success in the smartphone market difficult, and it has struggled to sell its 3G baseband outside of Apple,” adds Linley.

The analyst adds that in a quieter but significant move, Comsys acquisitions brings a processor for WiMax phones and was working on converting that design to support LTE.

Intel’s secret goal (well not anymore!) is to merge Infineon and Comsys technologies to create its own 4G solution coupled to its Atom chip for smartphones.

“Intel is clearly focused on smartphones, leaving the future of Infineon’s popular 2G processors in doubt. This deal could end up helping vendors such as Broadcom, MediaTek, and ST-Ericsson, which could step into the breach with their own 2G processors. Conversely, the deal gives Intel the technology it needs to develop an integrated 3G-smartphone processor, allowing it to compete against vendors such as Qualcomm and Marvell,” says Linley.

First published in TechPulse 360.


SeaMicro’s Minor Revolution In The Data Center

June 14, 2010

Not since the advent of virtualization has the data center faced such an opportunity for change.

Low-cost, ultra-low-power servers – sometimes called microservers – may finally get a jolt of legitimacy.

On Monday, secretive Santa Clara start-up SeaMicro formally launched its long-awaited remake of the x86 server: a 512-processor box that doesn’t use Intel’s ubiquitous Xeon chips but low-power Atoms instead. Atoms are the processors sold in $300 netbooks – giving rise to the observation that SeaMicro’s SM10000 is really just a collection of netbooks stuffed in a one box.

Expect vendors such as Dell to begin making microservers. SeaMicro claims it will cut power use by 75 percent.

It’s an observation that is essentially true. The result is a server that uses one-quarter the power and takes up one-quarter the space while performing the same amount of work. CEO Andrew Feldman says Atom is three times more efficient in performance per watt than Xeon. The reason is it can better power down when not in use and doesn’t waste energy trying to anticipate future workloads, as Xeon does.

That is why the SeaMicro box is better suited to the Internet, where traffic is bursty and generally only places lightweight demands on a server.

The SM10000 is the brainchild of Gary Lauterback, a former AMD fellow and Sun Microsystems engineer, and “is an enormous transformation of the data center,” claims Feldman.

He may not be exaggerating. Zeus Kerravala, a long-time tracker of the server industry at Yankee Group, says: “As an analyst I am often skeptical of technologies people tout as revolutionary, but this one I was really impress with.” If Dell and other top tier vendors aren’t already thinking about microservers “I’d be surprised,” he says.

In truth, SeaMicro isn’t the first company to conceive of low-power servers or ones running Atom. Super Micro Computer launched a rack-mounted Atom blade last year, and Hewlett-Packard markets a $400 MediaSmart home-server with Atom. In Austin, TX, Smooth Stone is working on technology to bring even lower-powered ARM processors, those in many cell phones, to the server market.

Improved efficiency is what motivated cloud-computing vendor Rackable Systems to make use of small servers with modest power to handle fluctuating Internet workloads.

However, SeaMicro hopes to take Atom boxes a step further. The company built into the SM10000 a 1.28-terabit communications fabric powerful enough for a super computer and installed a custom ASICs to handle the complex load balancing for 512 processors. A single box can replace 40 dual-socket, quad-core servers, two Ethernet switches and two terminal servers, says Feldman.

It also shrank the size of processor motherboards to the size of a credit card, taking off unnecessary components and reducing the power draw.

According to IDC, the package may catch on with Web 2.0 companies. Companies spend $27 billion globally a year buying energy to run their servers, the research firm says. Most would die to reduce the bill.

“I think it is a radical approach” that Web 2.0 companies will quickly adopt for their public clouds, says research analyst Katherine Broderick.

The SeaMicro, which raised $25 million from backers including Khosla Ventures, Draper Fisher Jurvetson and Crosslink Capital and received a $9.3 million Department of Energy grant, will make the box available in July. Selling for $139,000, it is likely to begin earning its investors a return.


GE Stands Behind WiMax In The Smart Grid, Sees Costs Dropping

April 12, 2010

Utilities were excited at first by the high-bandwidth, long haul capabilities of the wireless communications technology WiMax in the smart grid.

That excitement has waned. High costs and coverage gaps led many to favor alternatives for tasks other than simple “backhaul” data transmission to and from local collection points.

CE's Luke Clemente sees 4G technologies such as WiMax as part of the Smart Grid technology mix. "I'm very confident of that."

WiMax supporter General Electric seems to acknowledge this present, more limited role for the technology, long championed by chipmaker Intel and its distribution partner Clearwire. And yet, the smart-meter manufacturer continues to consider WiMax a viable alternative across the smart grid.

The winning smart-grid technology has not been picked, says Luke Clemente, general manager for metering and sensing systems. Data from WiMax smart grid trials are still coming in. GE last month announced its first WiMax smart-meter trial in the United States, after striking a deal with Consumers Energy in Michigan, and last fall kicked off a test of WiMax-enabled smart meters with SP AusNet in Australia and GE-funded partner Grid Net. WiMax smart meters are just being installed.

Utilities apparently continue to keep an open mind toward the technology, even as many favor private, proprietary networks. “We’re still finding customers that are discussing (WiMax) with us,” Clemente says . “They are waiting to see more data on performance.”

The biggest concern about the technology is its high cost compared with alternatives, such as radio frequency, or RF, mesh. RF mesh is found in meters from Landis-Gyr and Elster and used by communications companies Silver Spring Networks and Trilliant. Clemente argues that costs should fall as volume manufacturing begins and overhead is spread among a greater number of products.

He says GE also tells customers they get more bandwidth for their money. The public 3G technologies that telecommunications providers deployed were not cost effective for the smart grid, but the 4G technologies, such as WiMax and LTE, could emerge as more affordable and competitive.

“I see (WiMax) as a 4G technology (and) I think 4G will be part of the mix,” says Clemente. “I’m very confident of that.”


Router Company Highlights Failure Of Smart Grid Standards, Forms Independent Working Group

April 6, 2010

Smart-grid router company SmartSynch highlighted the failure of smart-grid standards Tuesday by forming an independent industry group to push for needed standards in communications gear.

Stephen Johnston said that despite three years of efforts, standards still are missing for smart meters and network equipment.

GridRouter maker Smartsynch hopes to push for IP-based communications standards on the smart grid. It has attracted a small group of supporters including AT&T and Motorola.

Now with utilities pouring billions of dollars of federal funds into early stage smart-grid efforts, projects risk of falling short from equipment that doesn’t interoperates or which can’t be easily installed.

Smart-grid standards initiatives date back more than three years. Yet smart meters from a company such as GE don’t work alongside one from Elster or Landis-Gyr, or connect smoothly to a GridRouter from SmartSynch.

“Without standards, we could be setting up the smart grid to be not as good as it could be,” Johnston said. “It’s a huge problem.” Prices also could be higher since competition among vendors is reduced.

On Tuesday, SmartSynch hopes to make a dent in the problem by forming an “ecosystem” group around its GridRouter. The group will push for interoperable, IP-based standards and has attracted several big name supporters: AT&T, Motorola, Rogers Communications, Itron, Cooper Power Systems and Battelle Energy. “We anticipate more,” Johnston said.

A day earlier, executives from Google, Intel, Honeywell, GE, AT&T and venture firm Kleiner Perkins Caufield & Byers made news with a letter to President Obama. They called on the Obama Administration to make the distribution of energy-use data to homes and businesses a priority.

But without standards, this could be more difficult than it seems. “If every utility deploys a proprietary system, those utilities can’t benefit from each other,” says Johnston.


Analyst: AMD 12-Cores vs. Intel 6-Cores Server Chips

April 1, 2010

This week, AMD and Intel launched their latest generation of high-end server chips.

The AMD Opteron 6000 (Magny-Cours) is a set of 8- and 12-core processors for dual- and quad-CPU servers. Not to be outdone, Intel has introduced the 8-core processor Xeon 7500 (Westmere), that could be deployed in mammoth 256-processor configurations.

To help us understand the differences between these 2 competing family of chips, we asked Insight64 analyst, Nathan Brookwod to share his views on these latest chips.

In 2-socket configurations, AMD delivers better performance despite Intel’s faster cores

AMD Magny-Cours (now the Opteron 6100) for both two socket and four socket configurations.

In the two-socket world, AMD faces Westmere-EP (Xeon 5600), which has six cores and twelve threads. Westmere’s cores are faster than Magny-Cours’ cores. Westmere wins some single-threaded contests, but AMD can overwhelm them if it gets to fire up all its cores, or if memory bandwidth or capacity is an issue.

In 4-socket configurations, AMD wins in price but Intel can scale to 64-sockets!

In the four-socket world, AMD faces Nehalem-EX, aka Becton, which has eight cores and sixteen threads. Becton’s cores are faster than Magny-Cours’ cores, and Intel has more on-chip cache (24MB to AMD’s 12MB) but AMD has 12 cores to Intel’s 8, and AMD charges the same price for the chips in 2P and 4P systems, while Intel charges almost 2x the price (per chip) for Becton, compared with Westmere-EP. Magny-Cours will win some benchmarks against Becton, and lose some, but it will always be less expensive. Some customers will value Intel’s greater expandability (to 8, 16 and 64-socket arrangements) and others will be attracted to AMD’s lower price.

Can Itanium survive to latest performance onslaught?

Brookwood’s law (“It’s easier to measure price than performance”) will give AMD an advantage in the two-way and four-way segments, but Becton will have almost no competition (other than Itanium and Power 7) for the higher-end niche of the server market. It’s hard to see how Itanium can survive for long against this Xeon onslaught.


[Video] Intel “convertible” Classmate PC 2.0: A Netbook For Kids

March 3, 2010

The next-generation of Intel's convertible Classmate PC is more powerful, larger, more rugged and more expensive

This week, Intel unveiled the second generation of its convertible netbook – which can be used as a traditional laptop in clamshell mode or folded into a tablet – for kids, the Classmate PC.

The new Classmate PC is a full-featured netbook with an Intel Atom chip, a hard drive or SSD storage, built-in wireless connectivity (Wi-Fi, WiMAX, 3G and GPS), audio/microphone/speakers and a rotational video camera; which could be used as an eReader with the bundled eBook application

Intel estimates the market for the classmate PC to be of a billion kids

In an exclusive interview with TechPulse 360, we asked Kapil Wadhera, the director of platform marketing at Intel for the Classmate PC to talk about the latest device, which will be available in a couple months and range from about $300-$500, depending on its configuration (wireless, storage, battery, screen resolution).

“The interest in this [the classmate PC] is worldwide because, if you look at the overall, there are about a billion kids in K-12 but the number of students that have access to technology or using technology as a tool one-on-one is very very small,” explains Wadhera.

The price for the current “clamshell” Classmate PC ranges from $200 to $400, while the “convertible” version commands a $30 to $50 premium, according to Intel representatives.

Follows is a video excerpt of our conversation with Wadhera.

We also had a chance to handle the Classmate PC and was impressed by its ruggedness with the rubberised coating, the larger waterproof keyboard and screen (10.1″ vs 8.9″).

However, the Classmate feels heavy – once you add the larger battery pack that is needed to achieve the 8.5 hours of battery life – and its touchscreen is not very responsive, which could be caused by the software that is still in beta, said Intel. More on the device in this video below.


[Video] Intel Confirms Launch Of “Westmere” Server Chip Mid-March, Details Cloud Security Functions

March 2, 2010

Intel server chip chief Kirk Skaugen confirmed the imminent release of the company's first 6-core chip since the ill-fated Dunnington

At a security event last night in San Francisco, Intel vice-president Kirk Skaugen confirmed the release date of Intel’s next-generation lineup of 32-nm Xeon server processors, including the first six-core Xeon chips since 2008 (Dunnington).

“In about 2 weeks it’s highly anticipated that we’ll be announcing this Westmere processor… When you buy that [chip], you should be able to get your return in about 5 months. But we’re probably most excited about – relative to just another crank of energy and performance – is the security features that are going into the processor,”told Skaugen to reporters during a media event hosted by security company RSA.

Scheduled in March 16, Intel will release a dozen dual-socket, 32nm Xeon processors as well a workstation version (Core i7) of the six-core chip.

“But what probably we’re most excited about – relative to another crank of energy and performance – is the [2] security features that are going into the processor,” added Skaugen.

The 2 new security features included in the Westmere line are:

  1. 7 new instructions call AES-NI, that will deliver encryption and decryption up to 9 times faster and up to 2 times more SSL functions than in the past;
  2. Trusted eXecution Technology (TXT) that is integrated in the processor, the chipset and Intel motherboards. “So for example, if you’re using VMotion to dynamically move a workload from a server to another, you want to set policies that say “I’m only going to move a workload to a server I know the secure root of trust has been verified… So, if I’m on a non trusted server I won’t be allowed to run [an application] on a trusted server, and vice-versa… all these kind of policies can be set up at the cloud level through the software that will be enabled on top of these new hardware,” said Skaugen.

Intel claims cloud domination

The Intel executive also provided reporters with some interesting statistics:

  1. a little under 40% of the world’s servers today are still single core;
  2. Xeon servers represent about 90% of the cloud infrastructure.

“With this new chip, everything that you know and love about Intel on energy efficiency, you’ll still get. So you can retire 15 single core servers and put in one Westmere server that is going to have the same performance, but you also going to get the trusted execution technology that can deliver that new secure root of trust,” concluded Skaugen.

Follows is a video excerpt of Skaugen comments:


[Video] RSA, VMware, Intel Unveil Trusted Cloud Vision

March 2, 2010

EMC collaborates with VMware and Intel to deliver proof of concept for business-critical security, compliance and control in the cloud

RSA Chief Art Coviella at the conference media reception

At a media reception tonight, RSA president Art Coviello unveiled a proof of concept for measuring and monitoring the security of a cloud infrastructure.

The concept that will make its debut tomorrow, at this week’s RSA Conference in San Francisco comprises of a hardware root of trust provided by Intel in its next-generation server processor (Westmere), a secure virtualisation environment and a security information and event management.

“What is more important, because this is about virtualisation (the enabling technology for the cloud infrastructure), is that we are building the security in [the cloud infrastructure] before we have massive deployments of cloud infrastructures. That was not happening 14 or 15 years ago. That, I think, is the biggest news that is coming out of this conference,” explains Coviello.

The goal is to:

  1. provide a better visibility into actual conditions within the bottom-most layers of the cloud, within physical and virtual machines, giving organizations the ability to verify secure conditions in what was formerly the “black box” of the cloud;
  2. enable finer controls to enforce differentiated policies in private clouds, such as what types of physical hardware virtual machines may run on and which tenants or business units may co-reside and share resources;
  3. streamline Compliance by providing automated processes for collecting, analyzing and reporting infrastructure-level activities and events.

Follows, a video excerpt of Coviello’s introductory comments.


Analyst: Imagination Technologies Powers Apple A4 Graphics Engine, Tops Nvidia Tegra

January 29, 2010

The A4 is Apple's custom chip powering the iPad tablet

[Update] Analyst Jon Peddie just blogged more details about iPad’s internals.

Inside Apple’s upcoming iPad tablet lies a mysterious chip: the A4, with the “A” presumably referring to Apple and the number “4″ perhaps to 4-cores!

First, here’s what we know for sure about the A4, either from common sense and public comments by Apple:

  1. It’s a system-on-a-chip (SoC) which combines a low-power ARM-based CPU (supposedly the latest Cortex A9), a graphics processor (or GPU), and other circuitries like audio and video codecs.
  2. Runs at 1GHz;
  3. Fabricated by Samsung.

In many aspects, Apple’s custom silicon is comparable to next-generation ARM-based SoC mobile processors from Freescale (i.MX series), Marvell (Armada), Qualcomm (SnapDragon) and Nvidia (Tegra); all showed at CES a number of tablets and netbooks prototypes using their respective chips.

Intel has also entered the SoC market with its Moorestown chip, a shrink of the current Atom processor used in netbooks, which now integrates a CPU and GPU on the same die.

If there’s not much mystery left on the origin of the A4 CPU – aside from the number of cores – little is still known however of the internal design of the chip and its graphics capabilities.

In a blog post that was widely reprinted all over the Net, Brightside suggested that Apple used ARM’s Mali 50 design for its core GPU. However, graphics expert Jon Peddie, of Jon Peddie Research, disagrees.

“The Ipad is not using Mali,” told Peddie to TechPulse 360. “The graphics engine in iPad certainly do, spec-wise, rival Tegra, Snapdragon, or Armada. On a polygons/second, or a pixel fill-rate basis it is as good as any out there, maybe better. And if game play is the criteria, then the graphics engine has all the power needed to deliver a very satisfying, if not impressive experience.”

But for Peddie, the real question should be: how does the A4 chip compares on a polygons/second/watt basis?

“And here is where you (and competitors) will be surprised – it will be top of the class. We won’t have measured data probably until late summer when all the tablets, including HP’s Slate and MSI’s Tegra-based unit are out and available for testing. But based on what we know about the engines involved, the A4 should be the best of breed,” added Peddie who is coming out next week with a full report on the A4 chip.

Peddie just blogged more details on the A4′s graphics engine: it’s an Imagination Technologies [PowerVR] SGX 535 core that is already used in iPhones and iPod touches.

Earlier this month at CES, Imagination unveiled a successor to the 535 core – the 545 – with even more impressive graphics capabilities. No doubt that it will find its way in next-generation iPhones, iPods and iPads.

Apple and Intel are key stockholders in Imagination Technologies. According to the UK publicly-listed company, they respectively own 9.5% and 16% (according to the Guardian) of its shares.


Video: Consumers Driving PC Market Growth, Intel says

December 17, 2009

Despite the recession, sales of consumer PCs grew in 2009. A trend expected to continue, at a double digit rate, through 2013

Intel has the consumer to thank for 2009 not being such a bad year after all.

“What we saw was a downdraft, especially on the commercial [PC] sector in 2009. But we actually saw unit volume growth for the consumer [PC] segment. That goes back to our point that PCs are really essential to the way we work and live and that the driver for the business in 2009 has been the consumer segment,” explains Intel vice-president Stephen Smith at a media briefing this morning in San Francisco.

Intel is now shipping in volume 32-nm chips of its Nehalem architecture (Core i3, i5 and i7) to OEMs; from 2 factories and plans to add 2 more next year. The Santa Clara, Calif.-based company will show off desktops and laptops using its new Core i3 and i5 chips, as well as netbooks at the Consumer Electronics Show (CES) next month.

“Looking forward, we see both the commercial and the consumer [segments] growing, with the consumer segment growing at a slightly faster rate in 2010… Again, if you step back 6 months ago, the idea that will would be talking about growth in the [consumer] segment in 2009 – and this is a double digit CAGR going forward – is a mindset people were questioning,” adds Smith.

Follows is a video excerpt of Smith’s pre-CES presentation.

Read the rest of this entry »


Follow

Get every new post delivered to your Inbox.

Join 32 other followers