HUBzero Paving the Way for the Third Pillar of Science

By Michael McLennan and Greg Kline

February 28, 2011

What if researchers could access and share scientific simulation and modeling tools as easily as YouTube videos with the power of the cloud to drive it all? That’s the underlying premise for the HUBzero Platform for Scientific Collaboration, a cyberinfrastructure developed at Purdue University.

HUBzero was created to support, an online community for the Network for Computational Nanotechnology (NCN), which the U.S .National Science Foundation has funded since 2002 to connect the theorists who develop simulation tools with the experimentalists and educators who might use them.

Since 2007, HUBzero’s use has expanded to support more than 30 hubs — and growing — in fields ranging from microelectromechanical systems and volcanology to translating lab discoveries into new medical treatments and the development of assistive technologies for people with disabilities.

HUBzero is now supported by a consortium including Purdue, Indiana, Clemson and Wisconsin. Researchers at Rice, the State University of New York system, the University of Connecticut and Notre Dame use hubs. Purdue offers a hub-building and -hosting service and the consortium also supports an open source release, allowing people to build and host their own. HUBbub2010, the first of planned annual HUBzero conferences, drew more than 100 people from 33 institutions as far away as Korea, South Africa and Quebec, along with U.S. universities nationwide.

Although they serve different communities, the hubs all support collaborative development and dissemination of computational models running in an infrastructure that leverages cloud computing resources and makes it easier to take advantage of them. Meanwhile, built-in social networking features akin to Facebook create communities of researchers, educators and practitioners in almost any field or subject matter and facilitate communication and collaboration, distribution of research results, training and education.

“Contributors can structure their material and upload it without an inordinate amount of handholding; that’s really a key because you want people to contribute,” says Purdue chemical engineering Professor Gintaras Reklaitis. He’s the principal investigator for, a National Science Foundation-supported Virtual Engineering Organization for advancing the science and engineering of pharmaceutical product development and manufacturing.

One could cobble some of this functionality together with commercial Web software, but HUBzero integrates everything in a single package. Add the research tool-enabling features and research-oriented functions like tracking the use of tools (useful for quantifying outreach) and citation tracking and you have something quite different — and powerful.

HUBzero can be a prime tool for satisfying cyberinfrastructure requirements, such as data management and access, of granting agencies like the NSF. HUBzero’s emphasis on interdisciplinary collaboration only makes it more attractive in funding proposals. A hub is central to the Purdue-based Network for Earthquake Engineering Simulation (NEES), a $105 Million NSF program announced in 2009, the largest single award in Purdue history. Purdue’s PRISM Center for micro-electro-mechanical systems and C3Bio biofuels research center, both funded by the U.S. Department of Energy, are some other recent major award winners employing hubs.

Such an infrastructure can have an impact on scientific discovery, as clearly shows.

As of December 2010, NCN identified 719 citations in the scientific literature that referenced In addition, user registration information indicates that more than 480 classes at more than 150 institutions have utilized nanoHUB. Because the site is completely open and notification of classroom usage is voluntary, the actual classroom usage undoubtedly exceeds these numbers. There are users in the top 50 U.S. universities (per the listing by U.S. News and World Report) and in 18 percent of the 7,073 U.S. institutions carrying the .edu extension in their domain name. “Nano” is a tiny area in science and technology, but nanoHUB is big in many institutions.

The community Web site now has more than 740 contributors and 195 interactive simulation tools. In 2010 more than 9,800 users ran 372,000 simulations. In addition to online simulations, the site offers 52 courses on various nano topics as well as 2,300 seminars and other resources, which have propelled the annual user numbers to more than 170,000 people in 172 countries.

Likewise, the cancer care engineering hub, one of the early hubs following nanoHUB, has proven to be the linchpin in building an online data tracking, access and statistical modeling community aimed at advancing cancer prevention and care.

“We were looking for a solution for sample tracking and data storage that would not cost $5 million and it was a true logistical challenge needing a comprehensive cyberinfrastructure support system,” says Julie Nagel, managing director of the Oncological Sciences Center in Purdue’s Discovery Park. “The hub is the core of the CCE project and has brought the project forward so much faster than we could have if we had started from scratch.”

The success of is what attracted the attention of Noha Gaber, who was seeking a good way to facilitate collaboration in the environmental modeling field when she came across the thriving international resource for nanotechnology research and education.

HUBzero, the technology powering nanoHUB, could obviously be used to build a Web-based repository of models and related documentation for projecting the spread and impact of pollutants. It also had built-in features, such as wiki space, enabling environmental researchers to share ideas and information. But the ability to make the models operable online, right in a Web browser window, and allow researchers to collaborate virtually in developing and using models was the deal closer.

“It’s not just providing a library of models, but providing direct access to these tools,” says Gaber, executive director of the U.S. Environmental Protection Agency’s Council for Regulatory Environmental Modeling. She’s a driving force behind the new for integrated environmental modeling.

Under the hood, HUBzero is a software stack developed by Purdue (and being refined continuously by the consortium and hub users) and designed to work with open source software supported by active developer communities. This includes Debian GNU/Linux, Apache HTTP Server, LDAP, MySQL, PHP, Joomla and OpenVZ.

HUBzero’s middleware hosts the live simulation tool sessions and makes it easy to connect the tools to supercomputing clusters and cloud computing infrastructure to solve large computational problems. HUBzero’s Rappture tool kit helps turn research codes written in C/C++, Fortran, Java, MATLAB, and other languages into graphical, Web-enabled applications.

On the surface, the simulation tools look like simple Java applets embedded within the browser window, but they’re actually running on cluster or cloud hosts and projected to the user’s browser using virtual network computing (VNC). Each tool runs in a restricted lightweight virtual environment implemented using OpenVZ, which carefully controls access to file systems, networking, and other server processes. A hub can direct jobs to national resources such as the TeraGrid, Open Science Grid and Purdue’s DiaGrid as well as other cloud-style systems. This delivers substantial computing power to thousands of end users without requiring, for example, that they log into a head node or fuss with proxy certificates.

The tools on each hub come not from the core development team but from hundreds of other researchers scattered throughout the world. HUBzero supports the workflow for all of these developers and has a content management system for tool publication. Developers receive access to a special HUBzero “workspace,” which is a Linux desktop running in a secure execution environment and accessed via a Web browser (like any other hub tool). There, they create and test their tools in the same execution environment as the published tools, with access to the same visualization cluster and cloud resources for testing. HUBzero can scale to support hundreds of independent tool development teams, each publishing, modifying, and republishing their tool dozens of times per year.

If a tool already has a GUI that runs under Linux, it can be deployed as-is in a matter of hours. If not, tool developers can use HUBzero’s Rappture toolkit to create a GUI with little effort. Rappture reads an XML description of the tool’s inputs and outputs and then automatically generates a GUI. The Rappture library supports approximately two dozen objects — including numbers, Boolean values, curves, meshes, scalar/vector fields, and molecules — which can be used to represent each tool’s inputs and outputs. The input and output values are accessed within a variety of programming languages via an Application Programming Interface (API). Rappture supports APIs for C/C++, Fortran, Java, MATLAB, Python, Perl, Ruby, and Tcl, so it can accommodate various modeling codes. The results from each run are loaded back into the GUI and displayed in a specialized viewer created for each output type. Viewers for molecules, scalar and vector fields, and other complex types can be automatically connected to a rendering farm for hardware-accelerated 3-D data views.

HUBzero sites also provide ways for colleagues to work together. For example, because of the unique way the HUBzero middleware hosts tool sessions, a single session can be shared among any number of people. A group of people can look at the same session at the same time and discuss ideas over the phone or instant messaging. If some of the people aren’t online or available, they can access the session later from their My Hub page and follow up at their convenience. Some commercial collaboration tools, such as Adobe Presenter, also work within HUBzero (hub builders are required to license these).

As people are using the tools, questions arise and sometimes things go wrong. HUBzero supports many ways for users to find help and help one another and includes a built-in trouble report system. Users also can post questions in a community forum modeled after’s Askville or Yahoo! Answers. In practice, many tickets aren’t really problems but are actually requests for new features. HUBzero supports a wish list capability for collecting, prioritizing and acting on such requests. Users can post an idea to the wish list associated with each tool or to the general list associated with the hub itself.

HUBzero’s unique blend of simulation power and social networking seems to resonate across engineering and science communities. As hub use continues to grow, a goal is to develop new capabilities to connect related content so that tools published on one hub can be easily found on all others. Another goal is to improve tool interconnection, so that one tool’s output can be used as input to another, letting developers solve larger problems by connecting a series of models from independent authors.

Michael McLennan is the senior research scientist and hub technology architect at Purdue. Greg Kline is the science and technology writer for Information Technology at Purdue (ITaP).

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!

Dell EMC will Build OzStar – Swinburne’s New Supercomputer to Study Gravity

August 16, 2017

Dell EMC announced yesterday it is building a new supercomputer – the OzStar – for the Swinburne University of Technology (Australia) in support the ARC Centre of Excellence for Gravitational Wave Discovery (OzGrav) Read more…

By John Russell

Microsoft Bolsters Azure With Cloud HPC Deal

August 15, 2017

Microsoft has acquired cloud computing software vendor Cycle Computing in a move designed to bring orchestration tools along with high-end computing access capabilities to the cloud. Terms of the acquisition were not Read more…

By George Leopold

HPE Ships Supercomputer to Space Station, Final Destination Mars

August 14, 2017

With a manned mission to Mars on the horizon, the demand for space-based supercomputing is at hand. Today HPE and NASA sent the first off-the-shelf HPC system into space aboard the SpaceX Dragon Spacecraft to explore if Read more…

By Tiffany Trader

HPE Extreme Performance Solutions

Leveraging Deep Learning for Fraud Detection

Advancements in computing technologies and the expanding use of e-commerce platforms have dramatically increased the risk of fraud for financial services companies and their customers. Read more…

AMD EPYC Video Takes Aim at Intel’s Broadwell

August 14, 2017

Let the benchmarking begin. Last week, AMD posted a YouTube video in which one of its EPYC-based systems outperformed a ‘comparable’ Intel Broadwell-based system on the STREAM benchmark and on a test case running ANS Read more…

By John Russell

Microsoft Bolsters Azure With Cloud HPC Deal

August 15, 2017

Microsoft has acquired cloud computing software vendor Cycle Computing in a move designed to bring orchestration tools along with high-end computing access capa Read more…

By George Leopold

HPE Ships Supercomputer to Space Station, Final Destination Mars

August 14, 2017

With a manned mission to Mars on the horizon, the demand for space-based supercomputing is at hand. Today HPE and NASA sent the first off-the-shelf HPC system i Read more…

By Tiffany Trader

AMD EPYC Video Takes Aim at Intel’s Broadwell

August 14, 2017

Let the benchmarking begin. Last week, AMD posted a YouTube video in which one of its EPYC-based systems outperformed a ‘comparable’ Intel Broadwell-based s Read more…

By John Russell

Deep Learning Thrives in Cancer Moonshot

August 8, 2017

The U.S. War on Cancer, certainly a worthy cause, is a collection of programs stretching back more than 40 years and abiding under many banners. The latest is t Read more…

By John Russell

IBM Raises the Bar for Distributed Deep Learning

August 8, 2017

IBM is announcing today an enhancement to its PowerAI software platform aimed at facilitating the practical scaling of AI models on today’s fastest GPUs. Scal Read more…

By Tiffany Trader

IBM Storage Breakthrough Paves Way for 330TB Tape Cartridges

August 3, 2017

IBM announced yesterday a new record for magnetic tape storage that it says will keep tape storage density on a Moore's law-like path far into the next decade. Read more…

By Tiffany Trader

AMD Stuffs a Petaflops of Machine Intelligence into 20-Node Rack

August 1, 2017

With its Radeon “Vega” Instinct datacenter GPUs and EPYC “Naples” server chips entering the market this summer, AMD has positioned itself for a two-head Read more…

By Tiffany Trader

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

Nvidia’s Mammoth Volta GPU Aims High for AI, HPC

May 10, 2017

At Nvidia's GPU Technology Conference (GTC17) in San Jose, Calif., this morning, CEO Jensen Huang announced the company's much-anticipated Volta architecture a Read more…

By Tiffany Trader

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

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

Quantum Bits: D-Wave and VW; Google Quantum Lab; IBM Expands Access

March 21, 2017

For a technology that’s usually characterized as far off and in a distant galaxy, quantum computing has been steadily picking up steam. Just how close real-wo Read more…

By John Russell

Russian Researchers Claim First Quantum-Safe Blockchain

May 25, 2017

The Russian Quantum Center today announced it has overcome the threat of quantum cryptography by creating the first quantum-safe blockchain, securing cryptocurrencies like Bitcoin, along with classified government communications and other sensitive digital transfers. Read more…

By Doug Black

HPC Compiler Company PathScale Seeks Life Raft

March 23, 2017

HPCwire has learned that HPC compiler company PathScale has fallen on difficult times and is asking the community for help or actively seeking a buyer for its a Read more…

By Tiffany Trader

Trump Budget Targets NIH, DOE, and EPA; No Mention of NSF

March 16, 2017

President Trump’s proposed U.S. fiscal 2018 budget issued today sharply cuts science spending while bolstering military spending as he promised during the cam Read more…

By John Russell

Leading Solution Providers

CPU-based Visualization Positions for Exascale Supercomputing

March 16, 2017

In this contributed perspective piece, Intel’s Jim Jeffers makes the case that CPU-based visualization is now widely adopted and as such is no longer a contrarian view, but is rather an exascale requirement. Read more…

By Jim Jeffers, Principal Engineer and Engineering Leader, Intel

Groq This: New AI Chips to Give GPUs a Run for Deep Learning Money

April 24, 2017

CPUs and GPUs, move over. Thanks to recent revelations surrounding Google’s new Tensor Processing Unit (TPU), the computing world appears to be on the cusp of Read more…

By Alex Woodie

Google Debuts TPU v2 and will Add to Google Cloud

May 25, 2017

Not long after stirring attention in the deep learning/AI community by revealing the details of its Tensor Processing Unit (TPU), Google last week announced the Read more…

By John Russell

MIT Mathematician Spins Up 220,000-Core Google Compute Cluster

April 21, 2017

On Thursday, Google announced that MIT math professor and computational number theorist Andrew V. Sutherland had set a record for the largest Google Compute Engine (GCE) job. Sutherland ran the massive mathematics workload on 220,000 GCE cores using preemptible virtual machine instances. Read more…

By Tiffany Trader

Six Exascale PathForward Vendors Selected; DoE Providing $258M

June 15, 2017

The much-anticipated PathForward awards for hardware R&D in support of the Exascale Computing Project were announced today with six vendors selected – AMD Read more…

By John Russell

Top500 Results: Latest List Trends and What’s in Store

June 19, 2017

Greetings from Frankfurt and the 2017 International Supercomputing Conference where the latest Top500 list has just been revealed. Although there were no major Read more…

By Tiffany Trader

IBM Clears Path to 5nm with Silicon Nanosheets

June 5, 2017

Two years since announcing the industry’s first 7nm node test chip, IBM and its research alliance partners GlobalFoundries and Samsung have developed a proces Read more…

By Tiffany Trader

Messina Update: The US Path to Exascale in 16 Slides

April 26, 2017

Paul Messina, director of the U.S. Exascale Computing Project, provided a wide-ranging review of ECP’s evolving plans last week at the HPC User Forum. Read more…

By John Russell

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