The Greening of HPC — A Conversation with Professor Wu Feng

By Caroline Connor, for HPCwire

November 18, 2010

Ever notice that everyone’s gone green these days? The increased awareness in the HPC community of the need to maximize energy efficiency in compute-intensive environments has never been greater. With The Green500 results coming out this week, my thoughts turned towards learning more about the man largely credited with the movement towards environmentally sustainable supercomputing, Professor Wu Feng from Virginia Tech.

HPCwire: Wu, you’re well known for your work establishing the need for green computing, starting with the “Green Destiny” system back in 2001 and more recently, establishing The Green500. What are the most significant changes you’ve observed in the direction of energy-efficient computing and its influence on supercomputer design?

Wu Feng: The single biggest change is simply the awareness that power and energy consumption is a problem in supercomputing. It’s like the four stages of recovery. The first step is recognizing you have a problem, then working to overcome it. It’s that simple.

HPCwire: The naming of “Green Destiny” seems self-explanatory. Now you’ve got a system called “HokieSpeed.” Where did you ever get a name like that for a supercomputer?

Wu: Unlike Green Destiny, where there were multitudes of meanings behind it, the name HokieSpeed is fairly transparent. I had to come up with a name quickly, kind of like when you talked to Thomas Sterling about his naming of Beowulf. For our project, we mainly have Virginia Tech “Hokie” investigators, who seek to accelerate their computational simulations to execute at high ‘speed’, so I just pushed two words together.

HPCwire: Everyone is talking about the Tianhe-1A supercomputer announcement. From what we know, does it look like it will rate well on The Green500?

Wu: We could see this coming around 18 months ago. China has made a strategic investment in HPC, and they seem to be putting their eggs into the GPU basket. I can’t comment right now on how the supercomputer rates on The Green500, but I will say that I would be surprised if it did not do well. From an energy efficiency perspective, we have discovered that GPU-based systems tend to be very energy efficient.

HPCwire: So, what are your thoughts on cloud computing with respect to energy efficiency?

Wu: To a large extent the jury is still out on this. It’s like “outsourcing” your IT support. Outsourcing has some benefits and some drawbacks.

With respect to energy efficiency, relative to an institution that is leveraging cloud computing, the advantage is they don’t have to worry about building the most energy-efficient machine room. That’s something that institutions are not necessarily experts at anyway, so why not just use a cloud?

I think of it as similar to being part of a timeshare. So, from the standpoint of energy use, timesharing on a cloud resource can be much more cost and energy efficient. Instead of owning private systems that are using only five percent of their capability, you can aggregate twenty users onto a single cloud resource that each uses five percent on a single resource instead of twenty. That ends up being more energy efficient from an aggregation standpoint. In fact, you’re increasing the efficiency of the whole ecosystem.

From a scientific computing standpoint, depending on how one defines scientific computing, it may not be as energy efficient for some users due to the potential necessity of first migrating large data sets up to the cloud. When work is completed it may then need to be brought down from the cloud for post-analysis, and moving data is very energy intensive. So while there isn’t a black-and-white solution to the problem, it’s the right solution for the right task at the right time.

HPCwire: Speaking of cloud, some of these huge Fortune 500s build their big computing sites near hydroelectric dams and wind farms. Do you study whether the energy used for supercomputing is from renewable sources, or perhaps non-carbon-emitting sources like nuclear? Do you envision a “LowCarbon500” as well as a “Green500” list?

Wu: I personally don’t do any research in the area of renewable resources, fuels or energy sources or what have you. I may be somewhat of a “Renaissance man” but not that much of one! (Laughing) I am already doing way too many things as it is!

I would hope there could be something like that, though I don’t know that The Green500 would ultimately be the resting place for it. I hope that what The Green500 does is raise awareness of the need for energy efficiency in the area of high-end computing. We are starting to do exploratory lists, like The Little Green500, and there have been discussions on creating additional exploratory lists that deal with other aspects of computation. For example, someone has suggested we do a cloud-based Green500 also. I think these are all wonderful ideas; but it’s more about trying to find the grassroots support and the mechanisms to move these projects forward.

HPCwire: So, how do you come up with your very best ideas?

Wu: A lot of my research is on building efficient systems, whether it’s efficiency with respect to performance, energy, cost or personnel. It permeates whatever I do. Since I am not quite as organized as I would like to be, I rely on using a variety of tricks or mechanisms to efficiently remember the ideas that are oftentimes the foundation for my research. Here’s an example:

During my time at Los Alamos National Laboratory, I used to road bike with a group of cyclists at lunchtime around a 25-mile loop. I discovered that cycling helped to free my inner consciousness and many of my deepest thoughts would simply “percolate to the top.” It was during these rides that I oftentimes had my best research ideas. My efficiency trick was to attach a tag to each idea and associate them with an acronym so I could remember them when I got back. Unfortunately my “physiological cache” was limited to only about eight ideas per ride. So, if I exceeded the eight (and remember, this is in the days before smart phones), I would often call my landline from my cell phone, while riding, and leave myself an “ideas message.”

HPCwire: Can you share where you grew up and some of your family background?

Wu: I had a rich and somewhat nomadic existence. I was born in Michigan and lived in maybe five to six places as a kid, 12 to 14 places in total. With all the moving that I did, everyone would say, “Oh you must be a military brat,” but that wasn’t exactly the case. In actuality, my dad was a professor of electrical and computer engineering. When he got the “right” overture from an institution, we would move and move frequently.

HPCwire: How do you believe that experience influenced the man you are today?

Wu: (Laughs) Ah, all of these Zen questions! Adaptability is certainly a very key element. When you’re exposed to as many different environments, cultures, ways of life, and points of view as I have, it enables you to understand different perspectives that are not necessarily in line with your own. It made me appreciate and respect different viewpoints, even if I did not agree with them. I learned to understand where people might be coming from, not to take anything too personally, and avoid making decisions that might be driven emotionally rather than logically. (Laughs) I am starting to feel like Spock on Star Trek now or something!

HPCwire: I discovered recently that you have a B.S. Honors degree in Music as an undergrad. Are you still a musician?

Wu: Yes. I was classically trained on the piano and picked up other instruments at school, the clarinet and the trumpet in particular. I think those were the sidelights that my parents encouraged, to keep me busy and out of trouble! (Laughing) My piano teacher was kind of bummed when I did not pursue music as my core major…though it did become a secondary major, an honors major. While I was good enough to place in national-level piano competitions as a teen, I felt that I wasn’t ever going to be good enough to make it on the international stage, which as a classical pianist, one would have to do to be successful. In retrospect, if I had been more disciplined I probably could have gone much farther. Hindsight is 20/20, but of course, I don’t have a crystal ball so I don’t really know what would have happened if I had chosen that path.

You know, it would be interesting to see how many of your column subjects are musically inclined, like John Gustafson and Michael Warren, for instance. I’ve met many such highly successful researchers who also have a musical background. Having such musical background exercises different parts of your brain and creates a holistic being. For example, in a study from the 1990s, researchers found that musical training and exposure strongly correlated to a substantial increase in size of the corpus callosum, the fibers that join the left and right hemispheres of the brain together. This in turn contributed to a larger amount of information relaying across different parts of the brain and pointed to music’s ability to aid in memory function. Anyway, enough of the geek talk…in short, I suspect you will find that those you view as being technically brilliant also possess a musical or otherwise creative background.

HPCwire: That’s an interesting observation. Wu, you mentioned earlier in this interview that you are “somewhat of a Renaissance man.” How do you think this came about?

Wu: It was around the time of Green Destiny and the huge amount of public exposure and interest the project generated, both inside and outside the HPC community. It created new collaborations and interdisciplinary research opportunities for me that expanded beyond my work in computer networking and influenced my exploring new areas of research. Someone said, “Oh, you’re a Renaissance man because you’re doing networking, green computing, now you’re doing bioinformatics.” And I said, “Yeah, that’s kind of the polite term for it.” (Laughing) From another perspective, one could say that I have a hard time focusing!

HPCwire: While researching for this interview, I came across your philanthropic project, MyVICE@VT, and your efforts to introduce computer science to rural and disadvantaged children at the Blacksburg New School. What was the catalyst that prompted you to take on yet another project when your plate is already pretty crowded?

Wu: I live in a neighborhood that has a lot of middle-schoolers, and being in a small community, you tend to get to know your neighbors both professionally and personally. I was seeing a trend with girls in particular that they were choosing not to go into what we call the STEM areas of education: Science, Technology, Engineering and Mathematics. We found that they were choosing not to go into these areas, not because they couldn’t do it, but because they perceived it as “not being cool.” What I tried to do with this project was to show them that it could be cool; you just have to package it in the right way. For example, at one rural school, the kids have an activity called Author’s Tea, where they write short stories, illustrate them, and bind them into books, which are then shared at an annual tea. I approached the kids with the idea that they could do their own movie adaptations of their short stories, like a Harry Potter movie. Before long, the kids were working with a system and programming with pictures without even realizing it. I wanted to show them that they could use these skills in whatever career path they chose to pursue without the fear of “not being cool.”

HPCwire: That brings me to ask, of all the diverse areas to which you have contributed your energy and talents, which do you find the most rewarding?

Wu: Succinctly speaking, the work I find most rewarding is that which crosses disciplinary boundaries. The work I’ve been doing in the bioinformatics area and work that builds bridges between computer science and biological/biomedical areas have been the most rewarding personally. I just wish I could spend more time on it. More broadly, the “motherhood and apple pie” answer is that what I find most rewarding is how what I do will ultimately benefit and contribute to society down the line. If I can influence change and help to make a difference that is more than rewarding enough for me.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

Researchers Scale COSMO Climate Code to 4888 GPUs on Piz Daint

October 17, 2017

Effective global climate simulation, sorely needed to anticipate and cope with global warming, has long been computationally challenging. Two of the major obstacles are the needed resolution and prolonged time to compute Read more…

By John Russell

Student Cluster Competition Coverage New Home

October 16, 2017

Hello computer sports fans! This is the first of many (many!) articles covering the world-wide phenomenon of Student Cluster Competitions. Finally, the Student Cluster Competition coverage has come to its natural home: H Read more…

By Dan Olds

UCSD Web-based Tool Tracking CA Wildfires Generates 1.5M Views

October 16, 2017

Tracking the wildfires raging in northern CA is an unpleasant but necessary part of guiding efforts to fight the fires and safely evacuate affected residents. One such tool – Firemap – is a web-based tool developed b Read more…

By John Russell

HPE Extreme Performance Solutions

Transforming Genomic Analytics with HPC-Accelerated Insights

Advancements in the field of genomics are revolutionizing our understanding of human biology, rapidly accelerating the discovery and treatment of genetic diseases, and dramatically improving human health. Read more…

Exascale Imperative: New Movie from HPE Makes a Compelling Case

October 13, 2017

Why is pursuing exascale computing so important? In a new video – Hewlett Packard Enterprise: Eighteen Zeros – four HPE executives, a prominent national lab HPC researcher, and HPCwire managing editor Tiffany Trader Read more…

By John Russell

Student Cluster Competition Coverage New Home

October 16, 2017

Hello computer sports fans! This is the first of many (many!) articles covering the world-wide phenomenon of Student Cluster Competitions. Finally, the Student Read more…

By Dan Olds

Intel Delivers 17-Qubit Quantum Chip to European Research Partner

October 10, 2017

On Tuesday, Intel delivered a 17-qubit superconducting test chip to research partner QuTech, the quantum research institute of Delft University of Technology (TU Delft) in the Netherlands. The announcement marks a major milestone in the 10-year, $50-million collaborative relationship with TU Delft and TNO, the Dutch Organization for Applied Research, to accelerate advancements in quantum computing. Read more…

By Tiffany Trader

Fujitsu Tapped to Build 37-Petaflops ABCI System for AIST

October 10, 2017

Fujitsu announced today it will build the long-planned AI Bridging Cloud Infrastructure (ABCI) which is set to become the fastest supercomputer system in Japan Read more…

By John Russell

HPC Chips – A Veritable Smorgasbord?

October 10, 2017

For the first time since AMD's ill-fated launch of Bulldozer the answer to the question, 'Which CPU will be in my next HPC system?' doesn't have to be 'Whichever variety of Intel Xeon E5 they are selling when we procure'. Read more…

By Dairsie Latimer

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Intel Debuts Programmable Acceleration Card

October 5, 2017

With a view toward supporting complex, data-intensive applications, such as AI inference, video streaming analytics, database acceleration and genomics, Intel i Read more…

By Doug Black

OLCF’s 200 Petaflops Summit Machine Still Slated for 2018 Start-up

October 3, 2017

The Department of Energy’s planned 200 petaflops Summit computer, which is currently being installed at Oak Ridge Leadership Computing Facility, is on track t Read more…

By John Russell

US Exascale Program – Some Additional Clarity

September 28, 2017

The last time we left the Department of Energy’s exascale computing program in July, things were looking very positive. Both the U.S. House and Senate had pas Read more…

By Alex R. Larzelere

How ‘Knights Mill’ Gets Its Deep Learning Flops

June 22, 2017

Intel, the subject of much speculation regarding the delayed, rewritten or potentially canceled “Aurora” contract (the Argonne Lab part of the CORAL “ Read more…

By Tiffany Trader

Reinders: “AVX-512 May Be a Hidden Gem” in Intel Xeon Scalable Processors

June 29, 2017

Imagine if we could use vector processing on something other than just floating point problems.  Today, GPUs and CPUs work tirelessly to accelerate algorithms Read more…

By James Reinders

NERSC Scales Scientific Deep Learning to 15 Petaflops

August 28, 2017

A collaborative effort between Intel, NERSC and Stanford has delivered the first 15-petaflops deep learning software running on HPC platforms and is, according Read more…

By Rob Farber

Oracle Layoffs Reportedly Hit SPARC and Solaris Hard

September 7, 2017

Oracle’s latest layoffs have many wondering if this is the end of the line for the SPARC processor and Solaris OS development. As reported by multiple sources Read more…

By John Russell

US Coalesces Plans for First Exascale Supercomputer: Aurora in 2021

September 27, 2017

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, in Arlington, Va., yesterday (Sept. 26), it was revealed that the "Aurora" supercompute Read more…

By Tiffany Trader

Google Releases Deeplearn.js to Further Democratize Machine Learning

August 17, 2017

Spreading the use of machine learning tools is one of the goals of Google’s PAIR (People + AI Research) initiative, which was introduced in early July. Last w Read more…

By John Russell

GlobalFoundries Puts Wind in AMD’s Sails with 12nm FinFET

September 24, 2017

From its annual tech conference last week (Sept. 20), where GlobalFoundries welcomed more than 600 semiconductor professionals (reaching the Santa Clara venue Read more…

By Tiffany Trader

Graphcore Readies Launch of 16nm Colossus-IPU Chip

July 20, 2017

A second $30 million funding round for U.K. AI chip developer Graphcore sets up the company to go to market with its “intelligent processing unit” (IPU) in Read more…

By Tiffany Trader

Leading Solution Providers

Amazon Debuts New AMD-based GPU Instances for Graphics Acceleration

September 12, 2017

Last week Amazon Web Services (AWS) streaming service, AppStream 2.0, introduced a new GPU instance called Graphics Design intended to accelerate graphics. The Read more…

By John Russell

Nvidia Responds to Google TPU Benchmarking

April 10, 2017

Nvidia highlights strengths of its newest GPU silicon in response to Google's report on the performance and energy advantages of its custom tensor processor. Read more…

By Tiffany Trader

EU Funds 20 Million Euro ARM+FPGA Exascale Project

September 7, 2017

At the Barcelona Supercomputer Centre on Wednesday (Sept. 6), 16 partners gathered to launch the EuroEXA project, which invests €20 million over three-and-a-half years into exascale-focused research and development. Led by the Horizon 2020 program, EuroEXA picks up the banner of a triad of partner projects — ExaNeSt, EcoScale and ExaNoDe — building on their work... Read more…

By Tiffany Trader

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Cray Moves to Acquire the Seagate ClusterStor Line

July 28, 2017

This week Cray announced that it is picking up Seagate's ClusterStor HPC storage array business for an undisclosed sum. "In short we're effectively transitioning the bulk of the ClusterStor product line to Cray," said CEO Peter Ungaro. Read more…

By Tiffany Trader

Intel Launches Software Tools to Ease FPGA Programming

September 5, 2017

Field Programmable Gate Arrays (FPGAs) have a reputation for being difficult to program, requiring expertise in specialty languages, like Verilog or VHDL. Easin Read more…

By Tiffany Trader

IBM Advances Web-based Quantum Programming

September 5, 2017

IBM Research is pairing its Jupyter-based Data Science Experience notebook environment with its cloud-based quantum computer, IBM Q, in hopes of encouraging a new class of entrepreneurial user to solve intractable problems that even exceed the capabilities of the best AI systems. Read more…

By Alex Woodie

Intel, NERSC and University Partners Launch New Big Data Center

August 17, 2017

A collaboration between the Department of Energy’s National Energy Research Scientific Computing Center (NERSC), Intel and five Intel Parallel Computing Cente Read more…

By Linda Barney

  • arrow
  • Click Here for More Headlines
  • arrow
Share This