Latest News On Computers Processors

Google

Monday, November 3, 2008

Intel: Spintronic CPUs at least 10 years away


Alternative methods of radically increasing the performance of CPUs, such as spintronics, will not find their way into production for at least another 10 years, Intel said this week in Taiwan.
Until now, computer processors have operated by using the binary on/off characteristic of electrons. However, the angular momentum of the electrons can be used, as well as the direction of that momentum, with the potential to increase the performance and capacity of chips.
However, over the next decade or so the focus will remain on incremental and more mundane improvements to get smaller and more efficient chips, according to Intel senior fellow and director communications technology lab, Kevin Kahn, who was speaking at the Intel Developer Forum in Taipei yesterday.
This is because the changes are better understood and relatively quantifiable as well as cheaper and more effective to deploy, Kahn said.
"We will push existing technology architectures as long and hard as we can because we know how to do them," he said, adding that a step-change such as using spin required a large amount of planning to be able to work with existing products: "No matter how hard you push on the new ones, it'll be a while."
The comments followed a futuristic keynote, in which Kahn detailed what humans could do to make computers smarter, perhaps heading toward a Terminator-like reality where computers outperformed their human counterparts.
He spoke about possible improvements in materials going into chips, as well as spintronics. He then moved onto sensors and actuators, displaying a robot which used an electronic field to sense when objects were near. The pre-touch had a shorter range than vision, but a longer range than touch, Kahn said.
The robot was able to take a dragon fruit without crushing it and hand it to Kahn. "You begin to believe that robot's alive," he said of the way it tracked his hand. "It's a very loose kind of following. It's like it's sniffing your hand out."
An actuation method he showed used sound. Sound, and not a fan, was generating the force that pushed a blue slip of paper upwards.
If receivers were built into mobile devices, and transmitters in specific spaces, whenever receivers were in range, they could automatically charge the devices, Kahn said.
The most unusual thing Kahn described was programmable matter, or small units that moved relative to one another to form objects, changing the form of devices or helping with computer simulations so operators can simply push or pull a design to make changes.
Current research in the area was using magnets or electrostatic forces, Kahn said, adding that, for him the idea was "in the space of magic".

Intel server revamp to follow AMD

Intel is getting ready to introduce a chip communications technology that mirrors an approach central to recent successes of rival Advanced Micro Devices.
If Intel's newly competitive chips recently brought to market act as the brains of a server, then the Common System Interface (CSI) is its nervous system. The technology, set for release in 2008, provides a new way for processors to communicate with each other and with the rest of a computer.
And alongside CSI, Intel plans to release an integrated memory controller, which is housed on the main processor rather than on a separate supporting chip. This will speed memory performance and so dovetail with the new communications system, the company expects.
Together, they could help Intel provide a much-needed counterpunch to AMD, which in 2003 introduced an integrated memory controller and a high-speed interconnect called HyperTransport in its Opteron and Athlon 64 processors. The two communication technologies, marketed together as "Direct Connect Architecture," deliver lower processor costs and chip performance advantages, which AMD has used to win a place in the designs of all of the big four server makers.
"Intel is hoping CSI will do for them in servers what 'CSI' did for CBS in ratings," said Insight 64 analyst Nathan Brookwood, referring to the hit TV series "CSI: Crime Scene Investigation."
Intel has been tight-lipped about CSI. However, Tom Kilroy, general manager of the company's Digital Enterprise Group, did confirm some details in a recent CNET News.com interview. Further glimpses have come from server makers, who are eager for CSI's debut in the "Tukwila" Itanium chip, due in 2008.
Tracking CSICSI brings two major changes. First, it will boost processor performance compared with Intel's current chip communication technology, the front-side bus.
"From a pure performance perspective, when we get to Tukwila and CSI, and we actually get some of the benefits of that protocol introduced into our systems, I think it's going to be really a big deal," said Rich Marcello, general manager of HP's Business Critical Server group.
CSI will be instrumental in helping double the performance of the Tukwila generation of servers, he noted.
Second, CSI will help Itanium server designers take advantage of mainstream Xeon server technology. Both chip families will use the interface, Kilroy said. That's particularly useful for companies such as Unisys, whose servers can use both processor types. It will make it possible for elements of a design to be used in both kinds of machine, reducing development costs and speeding development times.

Apple hires top IBM chip designer

One of IBM's top chip executives has agreed to join Apple as a senior executive, but he might have to fight off his former employer first.
Mark Papermaster, until recently IBM's vice president of microprocessor technology development, plans in to join Apple early November in a position that will see him working closely with Apple CEO Steve Jobs in what IBM believes is an attempt to expand Apple's presence in the markets for servers and chips for handheld devices, according to the copy of a lawsuit filed by IBM against Papermaster. IBM is suing Papermaster to prevent him from joining Apple and divulging trade secrets related to IBM's Power chips and server products, according to the complaint filed in the U.S. District Court for the Southern District of New York.
Papermaster has authored several papers on chip development at IBM, which of course used to make PowerPC processors for Apple before the company switched to Intel's processors in 2005. IBM called Papermaster "IBM's top expert in Power architecture and technology", and his most recent position involved managing IBM's blade server division.
An Apple representative declined to comment on the lawsuit or confirm Papermaster's pending employment with the company. IBM issued this statement: "Mr. Papermaster's employment by Apple is a violation of his agreement with IBM against working for a competitor should he leave IBM. We will vigorously pursue this case in court."
If Papermaster is able to successfully join Apple, he'll be working closely with Apple CEO Steve Jobs "providing to Apple technical and strategic advice on a variety of issues," according to IBM's complaint. But which issues?
Apple's Xserve servers have not exactly been a high priority over the last couple of years, as Apple has switched the Mac to Intel's processors and rolled out the iPhone. But a spruced-up Xserve blade server could be a nice complement to the Mac if Apple ever gets serious about tackling the enterprise market.
Still, Illuminata analyst Gordon Haff believes that Apple is unlikely to plunge back into the server market headlong after successfully pulling off the transition from a computer company to a consumer electronics company. Apple appeared to be serious about the server market when it launched the Xserve earlier this decade, but has spent less and less time extolling the product over the last two or three years, he said.
Papermaster's hire could signal Apple's intentions to build out a cloud-computing infrastructure to support things like MobileMe, or future services along those lines. Dense-but-powerful blade servers are being eyed by many companies as they build out the data centers of the future, and if Apple ever wants to be a major player in the future of Internet-delivered services, it is going to need a lot of computing power at its disposal. Papermaster's expertise in system design--putting together the entire package of processor, chipset, and the rest of the guts that form a computer--could serve him well at a company that prides itself on soup-to-nuts design.
As an extremely well-respected figure in the world of chip design, Papermaster might also be stepping in to lead Apple's chip design efforts. Apple's acquisition of P.A. Semi earlier this year showed the company is very serious about chip design. Jobs told The New York Times that P.A. Semi would be used to build chips for the iPhone and iPod Touch.
Papermaster's expertise lies with the Power architecture, of which Don Dobberpuhl's P.A. Semi team is also well-acquainted. The primary role for the Power architecture these days is in gaming consoles--all three major gaming consoles use a chip based on the Power architecture--but that does not necessarily mean Apple has that goal in mind, either.
If Apple wants to continue its strategy of designing and building complete systems, hardware, software, and now chips for iPhone and iPod Touch, it is going to need someone who can predict the future of chip design advise Jobs and Apple's executive team on how Apple can best take advantage of those trends. Papermaster, with a unique set of skills in the tech industry, might be just that guy. "They probably need somebody with an experience set that doesn't exist at Apple today," Haff said.

Intel Larrabee: An Introduction

If you haven't heard of Larrabee before reading this then on behalf of all of us, welcome back from whatever planet the aliens abducted you to. Intel's impending entry into the GPU market has been rumoured and talked about since late 2006, but up until now precious little concrete information has actually been revealed. The only solid information available was that Larrabee would comprise multiple x86 (Intel Architecture, IA) compliant cores thus making it capable of executing 'normal' code as well as running both the DirectX and OpenGL graphics-specific APIs.
With Intel's own developer forum, IDF, just around the corner and the more general, entirely graphics-centric, SIGGRAPH (Special Interest Group on GRAPHics and Interactive Techniques) conference starting on the 11th of this month Intel has finally allowed we, the general public, further information relating to Larrabee's architecture. The full paper details are being presented at SIGGRAPH and will appear at this link once available on the 12th of August.
Click to enlarge
Architecturally, Intel is making a massive departure from traditional GPU design with Larrabee. Annoyingly there's a lot of information on slides within the paper provided to us by Intel that we aren't able to publish (although we can at least write about said details) and even more that Intel won't talk about yet. Nonetheless, compared to the next-to-no information we've had up until now getting anything out of Intel is great.
Back in the early days of computer graphics, there was no such thing as a graphics card; everything was rendered in software by the CPU. As far as the average (if pretty loaded) consumer was concerned, that changed in 1996 with the release of the Voodoo 1 from 3dFX. For the fist time you didn't need to buy a new CPU (and usually, as a consequence, a whole new PC) to improve your gaming experience, you just slotted a PCB into in a PCI slot and away you went.
This move to discrete 3D graphics brought with it a divergence in architecture. CPUs carried on as they always had, optimising performance to carry out lots of different tasks pretty fast at the same time. GPUs, conversely, started to become much more fixed function and geared towards throughput performance. However, not all games are coded equally which means that in all but an ideal scenario, parts of the single-function card are sitting around waiting for something to do while other sections are overloaded with work.
This problem was partially solved in 2006, with the release of nVidia's G80 chip, the GeForce 8xxx-series and, a bit later, AMD's R600, 2xxx-series. The stream processor-based design upon which these and their successive GPUs operate means that some sections of that fixed function pipeline can be tailored on-the-fly to allow the GPU to adjust some parts of the rendering pipeline to best effect.


These trends towards massive parallelism and programmable, rather than fixed function processing lead the GPU's architecture on an interesting convergence with that of modern CPUs. Top end Nehalem chips will offer eight cores, each capable of processing two simultaneous threads and from what we hear from Intel, the future will see a tendency towards CPUs boasting many-core architectures, rather like a GPU.

Sun's largest investor ups stake

Sun Microsystems' largest shareholder, Southeastern Asset Management, has upped its stake and taken a more aggressive attitude toward its investment, indicating that it will take an active role in maximizing shareholder value through talks with management and third parties, according to a filing with the U.S. Securities and Exchange Commission on Wednesday.
Southeastern, an investment advisory service and self-described value investor in large-cap, mid-cap, and small-cap companies, increased its stake to 21.2 percent from its previous investment in September of 17.3 percent.
In noting the increased stake and change in its status as a passive investor to one that wants the flexibility to take a more active role, Southeastern stated in its filing:
"In this situation, Southeastern has talked to the issuer's management, and will have additional conversations with management and/or third parties, regarding opportunities to maximize the value of the company for all shareholders."
"To obtain the flexibility to discuss various alternatives, including any of the actions or transactions enumerated in clauses A through J of Item 4 of Schedule 13D, with the Issuer's management or with third parties, Southeastern is converting its ownership filing on Schedule 13G to a filing on Schedule 13D," Southeastern added.
Converting to a passive investment stake, as indicated in Schedule 13G, to a Schedule 13D allows Southeastern to take an active role in which it can work with other investors. Southeastern obviously is not a happy camper, especially after Sun's stock took a dramatic 17.5 percent plunge on an earnings warning Tuesday to close at US$4.77 a share.
Sun's stock is trading below US$5 a share for its second consecutive day. And in certain cases, institutional investors, such as pension funds and asset managers, are prohibited from owning stock in companies that trade for less than US$5 a share.
"We have talked to [Southeastern] and will consider their input, just like we consider the input from all our investors," said Dana Lengkeek, a Sun spokeswoman. She declined to comment on the content of those discussions with Southeastern, as well as with other institutional investors. She added Sun speaks to its investors on a regular basis.
According to SEC filings, Southeastern initially took a 10 percent investment stake in Sun Microsystems back in late May. That is when the stock was trading around US$13 a share.
Since then, Southeastern has steadily ramped up its investment as Sun's shares have continued to fall. In August, Southeastern noted in an SEC filing it had increased its stake to 16.5 percent and in Sun's proxy statement last month, it was listed as a 17.5 percent stakeholder.
A spokeswoman for Southeastern Asset declined to comment on the company's increased stake and desire to play an active role in maximizing shareholder value at Sun, noting that the company does not comment on companies in which it has holdings.
But in a BusinessWeek story published earlier this month, a Southeastern Asset manager had this to say about Sun Microsystems:

IBM and Intel® Drive Adoption of Open Switch Specification for Blade Servers


Switch Vendors Can Reach a Broader Set of Clients with a Single Product
IBM and Intel Corporation today announced they are extending their collaboration in the blade server market to drive adoption of an open industry specification for blade switches. This will enable switch vendors to maximize return on their research and development by developing products to a single design, reaching more customers with just one product.
Switches are essential components of every blade system, channeling data to and from the server. As part of the agreement announced today, IBM will extend the BladeCenter switch specification for blade servers to the Server Systems Infrastructure (SSI) organization on a royalty-free basis, allowing switch vendors to create one product that works across the BladeCenter and SSI ecosystems and driving more switch products for clients to choose from. The companies also announced plans to work together to establish a third-party Switch Compliance Lab, where the developer community can test and verify their blade server products for BladeCenter and SSI environments.
The announcement today builds upon a history of IBM and Intel collaboration on blade technology -- starting in 2002 with the development of the first BladeCenter servers and continuing in 2006 when IBM, Intel and other industry leaders formed Blade.org, the industry consortium driving open innovation in blade-based solutions that today has over 200 members.
SSI is an industry organization established over 10 years ago to enable server builders to develop compliant and interoperable building blocks for blade, chassis and manageability software technology. The incorporation of the open switch specification into SSI is intended to broaden the market for switch module vendors as they collaborate on SSI-specified blade switch solutions with Intel and IBM. In this segment, Blade Network Technologies, Brocade, NextIO and others have committed their support for the open specification. By extending its switch specification and ecosystem to affiliated members of SSI, IBM is driving greater adoption of BladeCenter-compatible switches.
"The extension of the BladeCenter switch specification and ecosystem to SSI advances open specifications for blade systems," says Alex Yost, vice president of IBM BladeCenter. "Making the switch design in our open BladeCenter specification available to a broader set of vendors demonstrates IBM's commitment to foster openness in the blade server market."
Demand for a common switch specification comes at a time when blade servers are becoming a critical ingredient in IT infrastructure. The availability of the open BladeCenter switch specification to SSI members allows them access to the more than 25 specification-compliant switches on the market today.
"By working with IBM, we have expanded ecosystem support for SSI blade specifications to over 70 vendors," says Kirk Skaugen, General Manager, Server Platforms Group, Intel. "Continued IBM and Intel collaboration will go further to expanding the entire blade server market. This is good for customers who will benefit from increased choice from a breadth of server, networking and storage vendors while enabling each product to maximize return on the R&D invested."
About IBM BladeCenterIBM has the broadest portfolio of blade offerings in the industry, including five blade enclosures, five compatible server blades and five I/O fabrics with a common architecture that allows clients to mix and match offerings. IBM has also taken a unique approach in the industry by opening up the IBM BladeCenter hardware architecture, which has created market and revenue opportunities for hundreds of companies selling peripherals like network and storage cards, switches and software. Since opening the BladeCenter specifications in 2004, 500 companies have downloaded the specifications for free. For more information about IBM BladeCenter, please visit: www.ibm.com/bladecenter.
About IntelIntel (NASDAQ: INTC), the world leader in silicon innovation, develops technologies, products and initiatives to continually advance how people work and live.

Intel Capital Announces Three New Investments in China Including its First Chinese Cleantech Investments

US$20 Million Investment in Trony Aims to Accelerate Renewable Energy Generation and Drive China's Sustainable Development

Intel Capital, Intel's global investment organization, announced today three new investments in China. This includes a new US$20 million investment in Trony Solar Holdings Co., Ltd., a leading Chinese thin-film solar energy and solution provider. Intel Capital also signed an agreement, subject to various closing conditions, to invest in NP Holdings Limited (NPH), a technology company focused on massive electricity storage systems. These are Intel Capital's first two cleantech investments in China and are intended to accelerate local green innovation and new ways to reduce dependency on carbon-based energy generation.
Intel Capital also signed an agreement to invest in Viewhigh, a leading software product and service provider for China's healthcare industry. Viewhigh focuses on healthcare business management, ranging from hospital operation and management to government supervision and financial management of healthcare institutions.
NPH is a cleantech company focused on electricity storage systems for renewable energy and energy efficiency. NPH strives to take the lead in the energy and environmental protection movement in China.
"These investments demonstrate Intel Capital's ongoing commitment to investing around the globe and its strategic focus on cleantech," said Arvind Sodhani, president of Intel Capital and executive vice president of Intel. "Clean technologies are transforming our world and the landscape of entrepreneurship."
Trony plans to use its funding, including the $20 million from Intel Capital, to enhance its production capacity to 105 megawatts,, to serve a broad base of customers, and strengthen its research and development capabilities. With growing R&D capacities, advanced application of innovative technologies, and low-cost volume manufacturing in Shenzhen, Trony now offers a diverse range of competitive products and solutions for various energy-related applications, such as solar farms in the utility industry and building materials for the construction industry.
"Intel Capital's investment will enable us to expand our leading position in the thin-film solar industry," said Li Yi, Chairman and CEO, Trony Solar Holdings Co., Ltd. "The expertise and network resources of Intel Capital consolidate our confidence to further advance in the energy sector."
"China's renewable energy industry is experiencing rapid development," said Cadol Cheung, managing director of Intel Capital Asia Pacific. "We believe these investments will be a catalyst to drive local cleantech innovation and help China toward the transition to a more sustainable energy system as well as economic growth."
Intel Capital is continuously looking for local start-ups and companies with strong potential for future growth. Over the past 10 years, Intel Capital has invested in more than 80 companies across mainland China and Hong Kong. Nearly 30 Chinese companies have received investments from Intel Capital's first US$200 million Intel Capital China Technology Fund. Intel established a US$500 million Intel Capital China Technology Fund II in April 2008. This fund aims to foster innovation and entrepreneurship while elevating the technology capabilities and global competitiveness of Chinese companies.