3Leaf Systems to Demo Silicon Enabled Server Virtualization Technology at Supercomputing 08

3Leaf Systems (www.3leafsystems.com), a provider of next-generation virtualization solutions for enterprise data centers, today announced that the company will showcase new silicon enabled virtualization technology enabling dynamic resource utilization using commodity servers from Super Micro Computer Inc. 3Leaf will demo the solution at Supermicro’s Booth# 1033 at SC08 being held in the Austin Convention Center, Austin, Texas from November 15-21.
Virtualization within high performance computing (HPC) environments presents particular challenges to today’s enterprises. HPC environments require ultra high speed message passing between applications in order to operate in high computing environments, and many of today’s virtualization solutions fall short in providing this capability. In addition, CPU and memory demands from HPC applications continue to drive over-provisioning of servers within the data center.

Now, rather than having to overbuild their infrastructures to allow for high compute or memory intensive jobs, and/or sacrificing performance to achieve virtualization, enterprises have another choice. 3Leaf’s silicon enabled technology demonstrates how enterprises can pool CPU and memory resources across multiple x86 servers to create a cloud environment, enable up to one terabyte of DRAM for memory intensive applications, and achieve low-latency messaging between applications.

“Virtualization, along with compute and memory constraints within today’s scale-out servers, provides many challenges for HPC environments,” said B.V. Jagadeesh, CEO and president, 3Leaf Systems. “With this technology, working together with companies like Supermicro, 3Leaf helps enable virtualization for HPC environments while lowering the capital and operational costs of their HPC infrastructures. Supermicro shares our vision and we are pleased the company is an early adopter of this revolutionary technology.”

“We recognize the strong, unique value that the 3Leaf technology achieves, and are excited to partner with 3Leaf to help enterprises overcome pain points in their HPC environments by optimizing resources and dramatically reducing data center costs,” said Charles Liang, CEO and president, Supermicro. “With the 3Leaf enabled Supermicro servers, enterprises can achieve a terabyte of DRAM at a dramatically low cost and enable low-latency messaging for fast performance.”

About SC08

SC08, sponsored by the ACM (Association for Computing Machinery) and the IEEE Computer Society Technical Committee on Scalable Computing and the IEEE Computer Society Technical Committee on Computer Architecture, will showcase how high performance computing, networking, storage and analysis lead to advances in research, education and commerce. This premiere international conference includes technical and education programs, workshops, tutorials, an exhibit area, demonstrations and hands-on learning. For more information, please visit http://sc08.supercomputing.org/.

Read More

rPath's Billy Marshall Interviewed on how Enterprises can Virtualize Applications as a Precursor to Cloud Computing

Billy Marshall, founder and chief strategy officer of rPath, recently spoke with Dana Gardner to find out more about how enterprises should begin moving to application-level virtualization that serves as an onramp to cloud benefits.

Many enterprises are factoring how to bring more applications into a virtual development and deployment environment to save on operating costs and to take advantage of service oriented architectures (SOA) and cloud computing models.

Finding proven deployment methods and governance for managing virtualized applications across a lifecycle is an essential ingredient in making SOA and cloud-computing approaches as productive as possible while avoiding risk and complexity. The goal is to avoid having to rewrite code in order for applications to work across multiple clouds -- public, private or hybrids.

The cloud forces the older notion of "write-once, run anywhere" into a new level of "deploy correctly so you can exploit the benefits of cloud choices and save a lot of money."

Here are some excerpts from that interview:


We're once again facing a similar situation now where enterprises are taking a very tough look at their data center expenditures and expansions that they're planning for the data center. ... The [economic downturn] is going to have folks looking very hard at large-scale outlays of capital for data centers.

I believe that will be a catalyst for folks to consider a variable-cost approach to using infrastructures or service, perhaps platform as a service (PaaS). All these things roll up under the notion of cloud.

Virtualization provides isolation for applications running their own logical server, their own virtual server. ... Virtualization gives you -- from a business perspective -- an opportunity to decouple the definition of the application from the system that it runs on. ... Then, at run-time, you can decide where you have capacity that best meets needs of the profile of an application.

I can begin sourcing infrastructure a little more dynamically, based upon the load that I see. Maybe I can spend less on the capital associated with my own data center, because with my application defined as this independent unit, separate from the physical infrastructure I'll be able to buy infrastructure on demand from Amazon, Rackspace, GoGrid, these folks who are now offering up these virtualized clouds of servers.

That's the architecture we're evolving toward. ... For legacy applications, there's not going to be much opportunity. [But] they may actually consider this for new applications that would get some level of benefit by being close to other services.

[If] I can define my application as a working unit, I may be able to choose between Amazon or my internal architecture that perhaps has a VMware basis, or a Rackspace, GoGrid, or BlueLock offering.

Read More

Intel Core i7 chip launches in Tokyo

Irasshaimase! Stores in Tokyo districts such as Akihabara have launched sales of Intel's Core i7 processor, due to be officially rolled out at U.S. resellers on Monday.

Computerworld reported that "several hundred people crowded stores" that opened around 10 p.m. Saturday. The top-of-the-line 965 chip sold out, according to one retailer.


Sofmap store in Akihabara is selling boxed Core i7 processors

(Credit: Sofmap)The Core i7 represents the vanguard of Intel's new Nehalem microarchitecture. The i7 is a desktop processor targeted initially at gaming boxes.

(See CNET review of Falcon Northwest Mach V tower system based on Core i7-965 processor.)

Sofmap, a large Japanese computer reseller, is showing Core i7 processors, motherboards, and systems on its Web site.

Sofmap lists the boxed Core i7-920 (2.66GHz) at 32,800 yen or close to $340. The i7-940 (2.93GHz) is listed at 63,800 yen or about $660. The high-end i7-965 (3.20GHZ) goes for 112,800 yen or $1,160.

A series of i7-920 processors bundled with Windows Vista Home Premium "DSP version" are shown ranging in price from 45,800 yen ($470) to 54,800 yen ($565).

Specification details (listed by Sofmap) include: LGA1366 socket, Quick Path Interconnect (QPI) rated at 4.8GT/sec (gigatransfers per second), 8MB L3 cache, and a 130 watt TDP (Thermal Design Power).

The DSP, or Delivery Service Provider, version of Windows Vista is being promoted heavily in Japan and is typically bundled with other components.


Sofmap advertisement for Core i7

(Credit: Sofmap)Motherboards and systems are also on sale. Motherboards are based on the Intel X58 chipset. DDR3 memory is being promoted along with the i7 too.

Tower systems using the Core i7 chip range in price from 179,800 yen ($1,852) to 219,800 yen ($2,264).

Read More

Where is Google voice-powered search?

Reporters were put into a frenzy this week when Google announced it was set to launch version two of its mobile search application for the iPhone that included the addition of voice-powered search, allowing you skip the keyboard altogether. But now the question is, where is it?

My colleague Josh Lowensohn reported on the application on Thursday, and duly noted on Friday afternoon that it still wasn't available in Apple's App Store. But as of Saturday afternoon, the application was still MIA. A search on the App Store returned only the older version of the Google Mobile App.


According to The New York Times, Google planned to release the free application through the iTunes Store "as soon as Friday." The application, an update to Google Mobile App, is meant to allow you to talk into your phone, ask any question, and the results of your query will then be offered up on your iPhone.

One reason for the delay could be that it has been bogged down by Apple's App Store approval process, which can take days or even months. Indeed, the Google Earth app for the iPhone took several days to appear in the App Store after its release. And Buzzd CEO Nihal Mehta noted that it took three months for his company's application to arrive in the App Store after it had been submitted. In other words, it's difficult for third-party developers to determine exactly when the application is going to be made available.

Perhaps from now on, when developers release an iPhone app, they'll learn to add a caveat that while the application has technically been released, it may take several days or even longer for it to actually show up in the App Store.

Read More

Adobe wants to bridge gap between PCs and cloud

Adobe Systems wants to have it both ways.

Microsoft's power with programmers is tethered to desktops and laptops, the vast majority of which run Windows. Google is trying to dominate what it believes is the new frontier, cloud computing, where applications run on the Web. Adobe, though, is trying to run down the middle with a strategy that touches on both domains.


"It's a balance of the client and cloud together that makes for the most effective applications and the best development," said Adobe Chief Technology Officer Kevin Lynch, who's planning to speak on the subject in a keynote speech Monday at the company's Max conference in San Francisco.



Adobe CTO Kevin Lynch

(Credit: Stephen Shankland/CNET News)
Since Adobe's $3.4 billion Macromedia acquisition in 2005, programming technology has been rising in importance within a company that got its start with publishing software such as Photoshop. The technology that brought the two companies together, Flash, will hog the spotlight at the conference.

Flash got its start as a way to give Web pages animations and basic applications such as games, but it's grown up since then. The Flex technology has given developers a more mature programming model, and the addition of video-streaming abilities to the Flash Player that's plugged into the vast majority of Web browsers has given Adobe's technology incumbent status. Who can live online without YouTube?

Adobe is still working on Flash, releasing Flash Player 10, aka Astro, in October. At Max, though, a Flash cousin called AIR--the Adobe Integrated Runtime--will share the stage with the release of version 1.5.

Flash and AIR are key to bridging the cloud-PC gap. For example, Adobe has launched an online Photoshop.com service, where members can upload, edit, and share photos. The site uses Flash to run the processing-intensive editing software on people's own computers, not Adobe's servers, Lynch said.

"Our operational costs for hosting that application are much lower than if we had server-side processing," and users get better performance, Lynch said.

But Flash still lives largely within the browser. Adobe hopes to uproot it with AIR, a "runtime" foundation for housing applications. AIR runs Flash programs but also has a built-in engine for showing Web pages and for running programs written in JavaScript, which is widely used for Web-based applications. And AIR is available for Windows, Mac OS X, and Linux, and programmers who write AIR applications don't have to worry about what operating system is on a person's computer.

But there's a risk to choosing a hybrid strategy: gains in flexibility often come at the expense of specialization, and specialized applications often work better. Sun Microsystems tried for years to get Java to catch on as a cross-platform runtime, but 13 years after its launch, it has yet to catch on with mainstream computing applications.

Google's cloud computing is limited by the comparatively feeble abilities of JavaScript running in Web browsers, but extensions such as Gears are bringing some advanced features. But Google gets the advantage of software that's available from all sorts of computing devices--your own, a friend's, a kiosk in an airport, your iPhone--as long as you have a network connection. And there's a natural collaboration component that comes with online applications that matches well with business needs.

Microsoft is moving slowly cloudward, but its cash cows remain Windows and Office. Its software is more powerful and responsive than any Web-based application--as long as you have your PC with you.

AIR applications can take advantage of local computing power, though--and the big new feature of AIR 1.5 is that it uses Flash Player 10, which brings 3D graphics, better text handling, the ability to mix different audio signals, and other abilities that make it a more reasonable competitor to Windows.

Read More