Once again, the world’s #1 event on accelerated computing will showcase scientific and engineering breakthroughs made possible by GPUs. The event runs March 18 to 21 in San Jose, Calif.

Featuring hundreds of talks and tutorials on everything from astrophysics to medical imaging and application development, GTC 2013 is the largest CUDA developer event. Thousands of the brightest minds from more than 40 countries will meet, network and share ideas.

For years, accelerated computing has been fueling the race for better science. Now, we’re seeing it advance the state-of-the-art in new commercial areas, and GTC will be the best place to see the latest in computing for big data analytics, computer vision and genomics/drug discovery.

It will also feature thought-provoking presentations about the profound impact of computational science on society. For example, world-renowned researcher and a featured speaker of TEDxBoston in 2011, Erez Lieberman Aiden, will deliver the Day Two keynote, on March 20.

Erez, who invented a method for three-dimensional genome sequencing, will discuss how his work can help unravel the human genome, unlocking the mysteries around genetic causes of disease and the environmental factors that impact genetic behavior (see illustration, left). You won’t want to miss it.

Elsewhere, you’ll find that GTC 2013 is packed with compute and developer-focused sessions on breakthroughs in science and industry across many fields.

From computational sciences to astronomy, genomics, energy exploration and more,  participants are sure to find inspiration for breakthroughs of their own.

The era of accelerated computing is here. Join us at GTC 2013.

Our CIO, Bob Worrall, was among a trio of panelists who spoke Monday at CES about how cloud computing is reshaping everyday devices.

“[The cloud] gives consumers more choice, especially in the Android gaming space, where consumers aren’t boxed into content from any one provider,” Worrall told a crowd gathered to learn how cloud computing is reshaping computer hardware.

The panel’s moderator, Michael Hickens, editor-in-chief of CIO Journal, noted that businesses will spend more than $100 billion on cloud-based services in 2016, up from $40 billion in 2013.

That spending is unlocking new business models that have already begun to reshape the electronics consumers use every day, as broadband connections to powerful banks of servers have made thinner, lighter devices possible.

Lars Fjeldsoe-Nielsen, from DropBox (left); OnLive CTO Tom Paquin (center); and NVIDIA CIO Bob Worrall (right). 

Worrall was joined on the panel by Lars Fjeldsoe-Nielsen, head of mobile business development at cloud-storage company DropBox, and Tom Paquin, CTO at online gaming provider OnLive.

“This is by far the most exciting time in my career,” Fjeldsoe-Nielsen said. “The impact that services have had over the past few years on the design of hardware has been absolutely incredible.”

The panelists spoke the day after NVIDIA introduced a new open-platform gaming device, codenamed Project SHIELD, and the NVIDIA GRID Cloud Gaming Platform, a powerful blend of hardware and specialized software that promises to do for games what Netflix did for video.

Such platforms will accelerate the trend towards cloud computing by letting companies focus on building new services, rather than the infrastructure needed to provide them, Paquin said.

“The ability to put more and more things in the cloud means that developers are going to be able to develop applications and systems and abilities that we have no concept of today,” he said. “This is good for consumers and opens up much more choice.”

The cloud opens up more choices for consumers, said OnLive CTO Tom Paquin (center).

All agreed that cloud computing creates new privacy and security challenges. But businesses like NVIDIA are finding new ways to utilize new kinds of connected devices. We’re working with industry leaders in hopes that as security solutions improve, we’ll be able to adopt some of the cloud-based services consumers now enjoy, Worrall said.

“If I lose my wife’s vacation photos, sorry dear; if I lose a chip design, that’s a different conversation,” he said.

While cloud computing has attracted wide attention over the past year, Worrall has been thinking about cloud computing for more than a decade.

Before NVIDIA, he worked for 20-plus years at Sun Microsystems – a company whose motto was once ‘the network is the computer’ – where he served as CIO.

This new GeForce 310.64 beta driver provides up to 38 percent faster gaming performance in Far Cry 3. The release is just the latest example of how NVIDIA works closely with game developers to tune its software to deliver the best performance for the year’s biggest gaming titles.

Two weeks ago, NVIDIA released a new GeForce R310 driver that delivered up to 26 percent faster performance for Call of Duty: Black Ops II and up to 18 percent faster performance for Assassin’s Creed III.

Download the new GeForce 310.64 beta drivers from GeForce.com.

GeForce 310.64 Driver Highlights

• Improves performance in Far Cry 3 by up to 38 percent¹.
• Plus, previously-released R310 optimizations:
• Up to 26 percent faster performance in Call of Duty: Black Ops 2;
• Up to 18 percent faster performance in Assassin’s Creed III;
• Smooth, shimmer-free graphics with TXAA anti-aliasing in Call of Duty: Black Ops and Assassin’s Creed III;
• Up to 16 percent in other top games likes Battlefield 3, The Elder Scrolls V: Skyrim and StarCraft II.

For more details, refer to the release highlights on the driver download pages and read the GeForce R310 article on GeForce.com.

NVIDIA strives to continually give gamers better performance and features with each driver release. Enjoy the new GeForce drivers and let us know what you think.

¹Measured on GeForce GTX 680 vs. GeForce 310.54 beta drivers

There are a lot of ways to measure Titan’s size. The machine is about as big as a basketball court. It contains 6,329 miles of interconnect cables. It’s cooled with 1,353 gallons of special refrigerant. Data is stored on 21,030 disks.

But the best way to understand Titan’s scale and complexity is to talk to one of the blue-coated engineers who scamper around its 200 towering black cabinets. Enger and his colleagues from supercomputer company Cray work under fluorescent lights as fans pump 1.3 million cubic feet of air per minute through the room. Ear plugs are recommended.

Mixing GPUs and CPUs makes Titan five times as efficient as its predecessor.

Two months ago pallets bearing the 18,688 NVIDIA Tesla GPUs that provide about 90% of the machine’s computing power began to arrive. That’s when Enger picked up his green and black power drill and got to work. It took Enger and 20 colleagues three weeks, working 7 days a week, to bolt all those GPUs into the machine.

The result could be the world’s most powerful computer. It won’t be official until November, when TOP500.Org releases its semi-annual list of 500 fastest supercomputers. But there can be no doubt Titan represents a breakthrough. At its peak, Titan cranks out more than 20 petaflops. That’s twenty thousand trillion floating point computations per second (‘floating point’ refers to a format many computers use to represent very small and very big numbers efficiently).

What’s really significant about Titan isn’t how many zeros you need to measure its performance, but how few megawatts Titan needs to do its work. Because it relies on GPUs to do much of the computing — rather than just CPUs — Titan requires only 9 megawatts of power.

Titan represents a step towards even faster ‘exascale,’ computing.

Titan is five times as efficient as Jaguar, the 2.3-petaflop computer it replaced at Oak Ridge. That efficiency comes thanks to an idea called ‘heterogeneous computing,’ says Buddy Bland, project director for the Oak Ridge Leadership Computing Facility.

“If this were a machine of the same power and it were using CPUs it would be using about 30 megawatts of power, or about $30 million a year,” says Bland. “So heterogeneous computing really gives us a lot more bang for the buck.”

That’s because GPUs rely on the parallel computing technology long prized by supercomputer engineers. In order to render virtual battlefields or imaginary dragons for video game enthusiasts, GPUs hustle through a number of tasks at the same time, rather than bouncing quickly from one task to another, as CPUs do.

It turns out that’s a very efficient way to do computing, says Bronson Messer, acting group leader for scientific computing at the Oak Ridge Leadership Computing Facility.

“The kind of physical things that happen in a game, it turns out those things happen in nature as well,” says Messer, who admits to knowing his way around a game controller. “These are exactly the kinds of problems we’re trying to solve in a lot of scientific questions, from combustion to climate.”

Some assembly required: Titan contains more than 18,000 GPUs.

The result is a sort of synergy between gaming and scientific research, with the tens of millions of consumers who rely on GPUs to power their games paying for research on a scale that the super-community could never afford on its own.

Yet the work done by those researchers is increasingly critical. Bland sees the simulations run by powerful machines such as Titan as playing an increasingly important role in scientific research. Titan is an open-science system, which means it can be used by researchers from academia, government labs, and private companies to model physical and biological systems ranging from the earth’s climate to the way engines burn fuel.

More powerful machines are coming. Titan – and its 18,688 GPUs — are a step forward on the path towards a concept Bland calls exascale computing. Titan can generate 20 thousand trillion flops. Exascale machines, by contrast, will generate one million trillion flops.

The U.S. Department of Energy would like to hit that mark by the end of the decade using just 20 megawatts of power. That’s a little more than twice what Titan consumes now.

Al Enger might want to start charging that power drill now.

Photos: Oak Ridge National Laboratory

Toyota Boshoku didn’t want to make engineers working on different machines, in different offices, work with separate copies of the same design. So they chose to use IBM’s SmartCloud Engineering Desktop.

IBM lets Toyota Boshoku give the same digital tools to engineers wherever they are, allowing them to work together on a single design housed on secure servers in a single data-center. That system is based on Citrix XenDesktop with HDX 3D Pro, running on NVIDIA’s GPUs.

The security benefits are obvious: engineers can access confidential plans without moving design data from the data center to local PCs. The real story here, however, is productivity: engineers working in far-flung factories can not just access their engineering data; they get flexible, powerful workstation capabilities wherever they go.

We’re working to accelerate this trend with the launch of the NVIDIA VGX K2 GPU. This new solution is built for the cloud. Based on NVIDIA’s power-efficient Kepler architecture, the VGX K2 will offer two of these Kepler GPUs on the K2 card, each with 4 GB of graphics memory, patented remote display technology to minimize lag and double the user density.

The new technology will be supported by Citrix’s CitrixReady partner ecosystem, an effort aimed at meeting the surging demand from companies such as Toyota Boshoku for systems that provide secure access to 3D design applications anytime, and anywhere.

That’s given rise to efforts to create a paper trail to certify these materials’ origins. The problem, of course: paperwork can be falsified.

Catherine McManus, chief scientist at Materialytics, has found a way to solve this problem using GPUs to help quickly verify the origins of a gem or a few ounces of raw material.

So far the 30-person company, based outside Austin, Texas, claims to be able to help identify the provenance of a wide range of different materials, including tourmaline, tin (cassiterite), coltan, diamonds, emeralds, gold, rubies, sapphires, and wolframite.

McManus starts with a technique known as laser-induced breakdown spectroscopy. The basic science can be understood by anyone who has taken high school physics: Materialytics fires a laser at the material it wants to examine, and then examines the scattered light created by the plasma that results to create a unique signature for the sample being examined.

Quadro GPUs help show differences between minerals gathered in different places.

From there, however, things get complicated: Materialytics measures each sample 64 separate times, collecting 40,000 data points each time, or more than 2.5 million data points per sample. That data piles up fast: a small scale test will gather more than 500 billion data points. Materialytics Chief Mathematician Jim Dowe says the company has accumulated terabytes of data on samples of raw material gathered from all over the world.

To make that data understandable, Materialytics relies on workstations equipped with NVIDIA Quadro GPUs. The company has created a matrix to display the data points associated with each sample along four different axes, giving users a way to visually compare and distinguish samples that may have been taken from different mines in the same region, or even different parts of the same mine.

Perhaps the biggest challenge: logistics. To match a sample with its origins precisely, the company has to start with materials from the original source. So far, the company says it has created created a collection of geological reference materials with more than 50,000 samples gathered from from 330 locations on all 7 continents.

“Getting into the areas where many of these mines are located is not simple,” McManus says. “We bring back tens and sometimes hundreds of pounds of rock.”

All that grunt work is beginning to pay off. Right now the early-stage startup is exploring a number of business models. Materialytics is exploring the idea of working with the electronics industry to verify the provenance of minerals such as coltan. It’s also talking to companies that trade in precious metals. The technology can even be used to help verify the authenticity of expensive manufactured goods – such as jet turbine blades and pharmaceuticals.

“The mineral itself certifies itself,” says Dowe. “The paperwork can be falsified along the way, but when it comes to the material itself, it cannot be anything but itself.”

To learn how people are making a difference with GPU computing, see the CUDA Spotlights page.

Prior to the launch of RemoteFX capabilities in Windows Server 2008 R2 SP1, a user of remote desktop protocol (RDP) was limited to software graphics only. Software graphics are rendered on the CPU, which wastes precious system resources and results in a low quality virtual desktop experience for end users. End users working in a virtual environment have had to settle with applications refreshing slowly and video stuttering. This degraded experience is simply no longer acceptable to those working in a virtual desktop environment. 

Now, by adding an NVIDIA GPU to servers hosting remote desktop sessions utilizing RemoteFX, users will see the kind of quality and performance for Microsoft Office productivity applications that traditional desktops and laptops have provided. To see these benefits, IT managers can choose servers either from Dell, HP, or IBM that support their choice of either a Quadro professional graphics card or Tesla M2070Q server graphics module. These servers powered by NVIDIA GPUs, and combined with RemoteFX, means end users can enjoy a truly ‘media rich’ experience. 

So what do I mean by “media rich?”  It’s realizing the true power that this VDI technology solution can bring to business users of all kinds.  [High-quality video conferencing for better collaboration, corporate trainings delivered over the web, Silverlight for interactive customer engagements, and even Windows Aero and Media Player—all being delivered on virtualized desktops, instead of traditional PCs or laptops. For more details on NVIDIA and RemoteFX, visit: http://www.nvidia.com/remotefx

If you’re in Atlanta this week at TechEd, be sure to visit us in the RemoteFX partner pavilion along with our OEM server partners, including Dell, HP, and IBM.  We’ll show how easy it is to utilize this.  With the capability to serve up to 24 VDI users per GPU, we, Microsoft, and our OEM hardware partners believe the enterprise IT budget savings of this virtualization solution are quite enticing.  Come by and see the performance, and discuss the potential ROI numbers with us for yourself.

TV on the PC is one of those promised advances that everyone anticipates and everyone wants. But until now, the solutions that enable you to tune into broadcast signals on your PC have simply been too expensive, too clunky, or too limited to really get traction with PC manufacturers . . . or consumers. Mirics Semiconductor of Fleet, UK, is changing that with a flexible and cost-effective solution using NVIDIA GPUs that gives your TV a passport to anywhere in the world.

Recently, Mirics CEO Simon Atkinson met with NVIDIA VP of Business Development Jeff Herbst to talk about the company’s breakthrough technology and explain how it works. (You can also learn more in this presentation.)

To get TV (or any broadcast signal, such as radio, for that matter) on your PC, you need an RF tuner to grab the signal you want and a demodulator to extract the information from that signal. The problem is, most PCs don’t have an integrated antenna, and, until now, no solution has worked globally in a way that’s cost-effective for manufacturers. TV systems vary by country or region, and each has its own quirks. To make a PC that could pick up FM radio in the U.S., broadcast TV in Europe or mobile digital TV in Brazil, you’d need to pack that PC with dedicated hardware demodulators for each and every signal. Not exactly an attractive proposition for OEMs.

Mirics solved this problem by eliminating hardware demodulators. It moved the demodulation function to software and it made that software run on a GPU. With CUDA-enabled GPUs, the algorithms that do real-time signal processing are offloaded from the CPU, leaving plenty of headroom for other applications.

The resulting product, Mirics FlexiTV, combines the RF tuner with the software demodulator. It’s inexpensive enough at a $5 bill of materials (versus almost $15 for an all-hardware solution) that it can be embedded in consumer hardware. And it works with any terrestrial broadcast signal. For PC manufacturers, it offers great differentiation for low-cost laptops, netbooks, and PCs. For consumers, it’s an excellent standard and HD TV viewing experience.

Ultimately, of course, mobile TV won’t be limited to PCs. Mirics is looking at Tegra systems, with an eye to pushing its offerings onto smartphones. The vision is to provide every device with the capability of receiving high-quality broadcast signals anywhere on the globe.

In fact, Mirics just announced its new FlexiStream home server, which allows any FlexiTV-equipped Windows 7 PC to view or record TV and then to stream that content to any portable device in the household, whatever its location.

Reading a hot best-seller is a serial process. Start at the beginning and read it to the end. But a task like counting the number of vowels in that same book can best be done as a parallel process. Give each paragraph to a different person, and it gets done far more quickly.

So it is with computing. Some tasks lend themselves to serial computing. But the complexity and data processing requirements that underlie our most challenging problems are rapidly moving beyond the capacity of serial processing.
We’ve all gotten used to Thomas Friedman’s idea that the world is flat. In solving problems with computers, we’ve similarly accepted the assumption that the world is serial.

In fact, the world is parallel.

Technology reflects the thinking in force at the time of its creation. And over time, it reflects our own self-imposed limitations. At some point, they have to be eclipsed.

Our previous computing approach, conceptually already over 40 years old, was to make single, serial CPU cores faster. Moore’s Law has enabled us to make faster, cheaper transistors but because of power constraints, it can no longer make single cores faster. All we can do to make CPUs faster is to add cores. But this guarantees diminishing returns — each new core can only process a small handful of threaded instructions. Everything must still be processed sequentially.

The practical effect – for an individual analyst, an individual piece of software or a multi-billion dollar research program – is effectively the same: Get in line.

Sequential processing is no longer adequate for the work before us now. The real question isn’t whether there’s some natural limit to Moore’s Law. It’s why we’ve allowed our progress to be limited by sequential CPUs, when massively-parallel GPU processors are already proving themselves orders of magnitude faster and cheaper.

Consider a range of distinct and (at first glance) entirely disconnected problems: 911 response time… dangerous weather patterns… breast cancer… national security… cleaner clothes… energy discovery… and financial derivatives valuation.

What these problems have in common is that a lack of computing speed impedes our ability to solve them. All of these issues – as disparate as they seem — have literally and demonstrably hit the limit of what’s possible with traditional, CPU-driven computers.

But all of them can be solved faster and cost-effectively with GPU machines, which have been proven to be hundreds of times faster than CPU clusters, at one-tenth the cost.

Consider what is now possible:

• Reducing 911 emergency response time – City planners and municipal response teams are combining datasets with physical mapping, population demographics, local resources, surface layers and vectors that involve many gigabytes of information in Geographical Information Systems. With GPUs, calculations that previously took 20 minutes to complete are now done in 30 seconds. What used to take 30 seconds is now done in real time.
• Predicting dangerous weather patterns – The most widely-used weather forecasting model in the world is “running out of gas for time-critical forecasts on conventional clusters,” according to its lead software developer at the National Center for Atmospheric Research. Adding more CPUs no longer improves speed. But NCAR says the effect of applying GPUs to the problem has been “transformative.” The result? More accurate and faster forecasting, critical for agencies around the world – particularly in regions in need of early warning, and those most likely to be affected by climate change.
• Fighting breast cancer – By replacing all 16 of its CPU clusters with two massively parallel (and far less expensive) Tesla GPU systems, an ultrasound process that once took three appointments can now be done in a single 30 minute appointment. This reduces anxiety, pain, and ultimately cancer incidence.
• Maintaining national security – GPUs form the basis for the most advanced tactical and strategic systems in the world. Seven GPU chips support the F-22 Raptor, The U.S. Air Force considers the F-22’s combination of stealth, speed, agility, precision and situational awareness unmatched — by any known or planned fighter.
• Keeping clothes cleaner – Researchers at Temple University are working to develop computer simulations that give companies like Proctor & Gamble a fast and cost-effective way to identify more effective and environmentally-sound detergents. Different chemicals attach themselves to different kinds of oils and soils more effectively. Traditionally, developing new detergents required extensive time and cost intensive testing in wet labs. Instead, the massive computational power provided by GPUs enables simulations of vast numbers of combinations, modeling the way different molecules attach themselves to (and banish) dirt.
• Securing energy supplies – With the search for energy becoming more complex and expensive, energy firms are constantly assessing massive quantities of seismic and geological data to determine the most efficient way to extract oil and gas and to maximize the utility of the reserves. A recent analysis of 740 square kilometers using 24 Tesla GPUs was completed 600 times faster than a traditional cluster of 66 CPUs — while using 95 percent less energy to run and cool the systems. Then consider improvements in automotive and transportation aerodynamics and fuel efficiency using GPU-powered design. Going parallel helps us discover energy, and conserve energy — and uses less energy to do both.
• Financial derivatives valuation –Recent market dynamics have brought even more focus on the need for accurate, predictive risk assessment models . Financial institutions can now use models that enlist the GPU’s massively parallel processing, to assess risk for a single trade or a portfolio accurately, with more confidence. Speed increases of 30X to over 100X mean that pricing a large portfolio of exotic swaps and derivatives can be handled in minutes instead of hours – supporting better decision-making and institutional stability.

All of these cases — like the code running on GPUs themselves – are just a small sampling of the array of problems that we can only solve time- and cost-effectively through parallel processing.

We are poised to make enormous strides in these domains, among countless others that demand the ability to process massive amounts of data quickly, inexpensively and accurately.

What GPU technology does – fundamentally – is bring simplicity to complexity. It helps us crack problems that, until recently, we simply couldn’t afford to solve, or couldn’t solve at all – whether it’s the design of a cardiac stent, a car body or a new molecule.

There’s a strong case that the only way civilization moves from one level to the next is by holding its earlier accomplishments with a loose grip.

CPU-based computing has served us well for decades. However, traditional sequential CPUs are not getting any faster, while our computational needs are growing exponentially.

The optimal – in fact, the only effective – computing that can take on the massive data problems we face is through massive parallel processing. And only GPUs can provide it.

On Monday, a leading developer of secure content management solutions, Kaspersky Lab in Russia, announced that they are using NVIDIA GPUs to improve the performance of their security software solutions.

And this isn’t a small increase in performance. Kaspersky Lab are seeing increases of 360 times as compared to a 2.6GHz Intel Core2 Duo.

Kaspersky Lab uses an NVIDIA Tesla S1070 1U GPU system to accelerate the screening of malicious programs using their own unique file similarity detection technology. When the Kaspersky anti-virus software on a computer suspects that a file may be malicious, even though it may not match any known virus signatures, the software uploads this file to the Kaspersky Lab data center. The server software then compares the suspected file against more than 50 million known good files and programs. Using complex anti-virus and SPAM detection algorithms, Kaspersky’s server software identifies the risk level of the suspected file and informs the client computer on what kind of preventive action to take. Pretty cool stuff!

Every day, more and more developers are finding awesome new uses for CUDA-enabled GPUs and are helping to transform every day computing for consumers.