VizWorld.com

IBM has come out with a new interactive visualization tool aimed at peeling back the many layers of bureaucracy in congress.  The new “Many Bills” systems lets you track bills as they make their way through the various committees and offices, each step along the way changing just a little bit, before becoming the bills you know.

Many Bills does a solid job of cataloging each Congressional bill (in every stage) with a color-coded format that creates an interlocking map allowing you to link themes and follow the path of different bits of legislation. It’s also a useful tool for those looking to delve into how Washington works, and see how the issues that matter to you are being drafted in our Capitol. It may even help expose politicians passing out legislative favors, and reward those staying true to the promises.

A great step forward for transparent democracy, now we just have to hope someone actually uses it.

via IBM’s Many Bills: This Law Visualization Tool Can Help Unlock Legislative Dealings | Fast Company.

Hardware congress, ibm, legal

IBM Research has a paper online from Bernice Rogowitz and Lloyd Treinish investigating the many strategies of colormaps used in data visualization, in hopes of creating a more stringent rule-based approach that can more intelligently apply default structures.  Included are many example visualizations like the one above.

Some of these ideas can be applied to more complex applications with multiple data sets in three dimensions, as illustrated in Figure 9.  These data are from an analysis of various weather observations, which indicate the state of the atmosphere on November 19, 1997 at 01:00 local time in the San Jose area.  Four distinct colormaps (two isomorphic and two segmented) are used to visualize four different variables using a variety of geometries registered into a single geographic scene.  The choice of colormaps used for each of these variables and their realizations is based upon their spatial characteristics and the task associated with the visualization.  For example, relatively noisy data such as wind speed are primarily mapped into luminance, while relatively smoothly varying data such as temperature are primarily mapped into opposing saturation pairs to impart a continuous representation. When contouring is selected as a technique, the data are mapped into a set of bands, to which a segmented colormap with perceived ordinality is applied.

via Why Should Engineers and Scientists Be Worried About Color?.

Science ibm, research

In a moment reminiscent of the SGI/PSC debacle, IBM has just announced that they are pulling out of the NCSA & University of Illinois “Blue Waters” project, begun back in 2007 to create a sustained-Petascale computer.

The University of Illinois and NCSA selected IBM in 2007 as the supercomputer vendor for the Blue Waters project based on projections of future technology development. The innovative technology that IBM ultimately developed was more complex and required significantly increased financial and technical support by IBM beyond its original expectations. NCSA and IBM worked closely on various proposals to retain IBMs participation in the project but could not come to a mutually agreed-on plan concerning the path forward.IBM will return money received to date and NCSA will return equipment delivered by IBM per terms of the contract.

Petascale computers aren’t quite the “big deal” anymore, as there are already a handful of them on the Top500.  I assume the disagreement was on how to proceed forward to an Exascale computer.

via Blue Waters computing system.

Hardware ibm, ncsa

IBM has just announced their next “Five in Five”, five innovations that they believe will change the world in just 5 years.  Starting with “ubiquitous sensors”, I particularly like their belief of 3D in five years.  Not in the movies or TV, but in User Interfaces and Video Chat.

In the next five years, 3-D interfaces – like those in the movies – will let you interact with 3-D holograms of your friends in real time. Movies and TVs are already moving to 3-D, and as 3-D and holographic cameras get more sophisticated and miniaturized to fit into cell phones, you will be able to interact with photos, browse the Web and chat with your friends in entirely new ways.

Watch the video for all 5 innovations, presented with some nice CG to demonstrate what they think will happen.

via IBM Press room – 2010-12-27 IBM Reveals Five Innovations That Will Change Our Lives in the Next Five Years – United States.

Science ibm

IBM and the Multimedia Development Corporation (MDeC) have just completed installation of the ‘MAC3 Rendering Farm”, a cloud rendering service that is part of the MSC Malaysia Animation and Crective Content Center.  Hoping to turn Malaysia into a hub of visual effects artists and animators by providing centralized remotely-available rendering services, the cloud offers an impressive 8x performance boost over local workstations.

“It is a well-known fact that rendering is expensive and the MAC3 high performance rendering facility will help local creative content companies, especially the start-ups, to overcome the cost barrier,” said Deputy Prime Minister Deputy Prime Minister Tan Sri Muhyiddin Yassin. “With this facility in place, Malaysia’s creative content companies will have the technological edge in their rendering and production processes. At the same time, it will also help them lower their costs,” he said.

To get access, simply follow the directions on their website .  The system seems to be a 128-node Dual-Quad-Core Intel systems with 80TB of storage, allocating 1.5TB per user.

via IBM Builds Animation Cloud for Multimedia Development Corporation | CloudTweaks.com – Cloud Computing Resources.

Hardware ibm, renderfarm

A research paper from IBM analyzes a FLOP-Intensive algorithm (dot-products and additions of 2D matrices of single-precision floating point values), and finds that the CPU can actually beat the GPU versions.  I’ll let the abstract do the talking (although I’ve reformatted it for readability).

We implement this algorithm on a nVidia GTX 285 GPU using CUDA, and also parallelize it for the Intel Xeon (Nehalem) and IBM Power7 processors, using both manual and automatic techniques. Pthreads and OpenMP with SSE and VSX vector intrinsics are used for the manually parallelized version, while a state-of-the-art optimization framework based on the polyhedral model is used for automatic compiler parallelization and optimization.

The performance of this algorithm on the nVidia GPU suffers from:

1. a smaller shared memory,
2. unaligned device memory access patterns,
3. expensive atomic operations, and
4. weaker single-thread performance.

On commodity multi-core processors, the application dataset is small enough to fit in caches, and when parallelized using a combination of task and short-vector data parallelism (via SSE/VSX) or through fully automatic optimization from the compiler, the application matches or beats the performance of the GPU version.

The primary reasons for better multi-core performance include larger and faster caches, higher clock frequency, higher on-chip memory bandwidth, and better compiler optimization and support for parallelization. The best performing versions on the Power7, Nehalem, and GTX 285 run in 1.02s, 1.82s, and 1.75s, respectively. These results conclusively demonstrate that, under certain conditions, it is possible for a FLOP-intensive structured application running on a multi-core processor to match or even beat the performance of an equivalent GPU version.

Not really surprised to see the Power7 perform the best (Given that it’s IBM’s chip and IBM engineers at the helm here).  If you read the paper, you’ll see that they use a 500×500 image (4MB in size) with matrices that require 250KB of space (page 6).  This won’t fit into the cache of the GTX285 so they spend much time paging data in and out.

I’ld be very curious to see if the new Fermi GTX480 changes this any.

via IBM Research | Technical Paper Search | Believe it or Not! Multicore CPUs can Match GPUs for FLOP-intensive Applications!|(Search Reports).

Science algorithm, ibm, paper

HPC people have been investigating Tesla & other CUDA technologies for several years, but mainly through test & development servers and homebrew configurations.  No longer will that be required, as now IBM is jumping in the game by packing the new Tesla M2050 card into their new systems.

“NVIDIA provides an innovative solution for customers who push the envelope in high-performance computing,” said Dave Turek, vice president, Deep Computing, IBM. “GPU acceleration provides performance boosts for many applications in energy exploration, science and financial services. It is among the significant emerging supercomputer technologies to watch in the years ahead.”

The industry standard benchmarking tool LINPACK boasts an 8x improvement when run on a server with 2 M2050′s installed, so the Top500 might see some amazing upsets from smaller IBM clusters this year.

Full release after the break.

Read more…

Hardware gpgpu, ibm, linpack, nvidia, tesla

If you’ve followed IBM’s Many Eyes and the IBM Visual Communication Lab very closely, then you’ve probably heard the names Fernanda Viegas and Martin Wattenberg.  They recently left Big Blue and formed their own company named “Flowing Media”.  They sit down with InformationAesthetics in a great interview on why they moved and what they hope to do in their new business.

Why did you start Flowing Media?

We believe that visualization is ready to come of age as a communication medium, and we’re excited to focus full-time on consumer and mass-audience visualizations.

We see a huge range of applications for this flavor of visualization. A non-profit group might want to draw widespread attention to data on the environment. A news organization might want a new set of tools for its reporters. A fashion house might even see the chance to make a striking statement.

Flowing Media offers strategy, design, and development services. We can help figure out what kind of visualization is right for a particular purpose, and then invent and design the technology to bring data to life.

via Interview: Fernanda Viégas and Martin Wattenberg from Flowing Media – information aesthetics.

Science ibm, interview

From IBM Research in Zurich:

IBM scientists have created the smallest 3D map of the earth – so small that 1,000 maps could fit on a grain of salt. The scientists accomplished this through a new, breakthrough technique that uses a tiny, silicon tip with a sharp apex — one million times smaller than an ant — to create patterns and structures as small as 15 nanometers at greatly reduced cost and complexity. This patterning technique opens new prospects for developing nanosized objects in fields such as electronics, future chip technology, medicine, life sciences, and opto-electronics.

Science ibm, maps, nanotechnology

Motion Theory follows up their interesting IBM Data Baby commercial with two new offerings, Data Transportation and Data Energy.

The new spots continue to build on the innovative way IBM gathers and analyzes information and utilizes it to create efficiencies in everyday life. They are striking in different ways – “Energy” features beautiful holographic images flowing from energy sources while “Transportation” is an abstract visual metaphor for how people move around the world.

See both spots below.

Graphics commercial, ibm, motion theory