NanoHUB@Home

nanoHUB@Home
nanoHUB@Home logo
Project
Status	Inactive
Category	Nanotechnology, Nanoscience
Compute	CPU
Requires	VirtualBox
Development
Developer	Purdue University
Sponsor	Network for Computational Nanotechnology
Maintainer	nanoHUB team
Initial release	September 1, 2019  (7 years ago)
Discontinued	March 30, 2024  (2 years ago)
Software
Written in	C++, BOINC
Operating system	Windows, Linux, macOS
Size	Variable virtual machine workloads
BOINC statistics
Stats as of	June 18, 2021  (5 years ago)
Active users	0
Total users	705
Active hosts	0
Total hosts	1,288
Metadata
Website	https://boinc.nanohub.org/nanoHUB_at_home/
License	Mixed; primarily open-source scientific software

[[File:{{#setmainimage:Nanohub.png}}|alt=logo image|center|frameless]]

nanoHUB@Home (Wayback Machine snapshot, 2024-03-30) was a BOINC-based volunteer computing project operated through nanoHUB, an online science and engineering gateway focused on nanoscience and nanotechnology. The project allowed volunteers worldwide to donate unused CPU resources to assist with computational nanotechnology simulations and machine learning research.

The project was connected to the broader nanoHUB cyberinfrastructure developed by the Network for Computational Nanotechnology (NCN) at Purdue University. nanoHUB became one of the world's largest scientific gateways dedicated to nanotechnology research and education, serving researchers, educators, and students in more than 170 countries.

Why nanoHUB@Home?

Nanotechnology research often requires large-scale computational simulations involving quantum mechanics, materials science, semiconductor physics, molecular dynamics, and nanoscale device modeling. Many of these calculations are computationally intensive and can take hours or days on traditional systems.

nanoHUB@Home was created to supplement the computing resources available to nanoHUB users by leveraging distributed volunteer computing through BOINC. Instead of relying solely on centralized clusters or cloud resources, simulations could be distributed across thousands of volunteer computers around the world.

The project also supported experimental machine learning efforts designed to generate predictive models from large collections of simulation results. According to Purdue researchers, this enabled approximate simulation results to be generated rapidly while reducing the need to rerun expensive calculations repeatedly.

Goal

The primary goal of nanoHUB@Home was to connect volunteer computing to the simulation needs of nanoHUB, an online science gateway focused on nanotechnology. nanoHUB is operated by the Network for Computational Nanotechnology, headquartered at Purdue University.

The project aimed to:

• Accelerate nanotechnology simulations
• Expand computational resources available to researchers
• Improve educational access to simulation tools
• Support machine learning and predictive modeling research
• Demonstrate the usefulness of volunteer computing for computational nanoscience

Methods

nanoHUB@Home used the BOINC middleware platform developed at the University of California, Berkeley. BOINC distributes computational work units to volunteers and validates returned results using redundancy and cross-checking techniques.

Unlike many traditional BOINC projects, nanoHUB@Home required VirtualBox in addition to the BOINC client.

Virtual machines were used because nanoHUB simulation tools often required complex scientific software environments and dependencies that would have been difficult to deploy natively across many operating systems and hardware combinations.

The project supported more than 200 simulation tools deployed through nanoHUB.org. These applications covered a broad range of topics including:

• Semiconductor device simulation
• Quantum transport
• Nanomaterials
• Nanoelectronics
• Molecular modeling
• Materials engineering

Workloads ranged from short simulations lasting only minutes to large computational jobs running for several hours depending on the scientific model and parameter complexity.

The distributed infrastructure enabled volunteers to process workloads generated by educational and research users of nanoHUB. Project developers reported that volunteer computers consumed work units at a rate "at least one order of magnitude higher" than other nanoHUB computing venues, requiring multiple infrastructure and database upgrades during 2019.

The project also contributed toward building machine learning datasets capable of interpolating simulation results across parameter spaces.

History

The underlying nanoHUB platform evolved from earlier Purdue University initiatives dating back to the 1990s. The original PUNCH project later evolved into nanoHUB, which officially launched in 1998.

In 2002, the National Science Foundation funded the Network for Computational Nanotechnology to expand nanoHUB into a major scientific gateway.

nanoHUB@Home itself launched publicly in 2019. Early project news updates documented rapid consumption of work units by volunteers and infrastructure upgrades required to keep pace with demand.

Archived server status pages indicate that the project became inactive around 2021.

Project team / Sponsors

The project was operated primarily through Purdue University and the Network for Computational Nanotechnology.

Notable contributors included:

• Gerhard Klimeck — Director of NCN and nanoHUB leader
• Alejandro Strachan — Materials engineering researcher and machine learning lead
• Ben Haley — nanoHUB software engineer
• Michael Zentner
• Steve Clark
• Claire Stirm

nanoHUB and NCN received major support from the National Science Foundation.

Technology

nanoHUB@Home relied on several major technologies:

• BOINC distributed computing middleware
• VirtualBox virtualization
• HUBzero cyberinfrastructure platform
• Scientific simulation software hosted through nanoHUB
• Machine learning workflows for simulation analysis

The virtualization approach used by the project resembled concepts explored in academic work surrounding virtualized BOINC environments.

Scientific impact

nanoHUB became one of the world's largest nanotechnology gateways and accumulated thousands of scientific citations over its operational lifetime.

The broader nanoHUB ecosystem provided:

• Hundreds of simulation tools
• Thousands of educational resources
• Large-scale classroom integration
• Collaborative online research environments
• Cloud-based scientific workflows

Researchers described nanoHUB as one of the earliest broadly successful scientific cloud computing environments.

Scientific publications

• Gesing, Sandra, Michael Zentner, Steve Clark, Claire Stirm and Ben Haley. HUBzero®: Novel Concepts Applied to Established Computing Infrastructures to Address Communities' Needs. PEARC '19: Practice and Experience in Advanced Research Computing (2019). DOI: 10.1145/3332186.3332238.

• Anderson, David P. BOINC: A Platform for Volunteer Computing. Journal of Grid Computing (2020). DOI: 10.1007/s10723-020-09521-x.

• Klimeck, Gerhard et al. nanoHUB.org: Advancing Education and Research in Nanotechnology. Computing in Science & Engineering (2008).

Legacy

nanoHUB@Home demonstrated that volunteer computing could be applied effectively to computational nanotechnology workloads, particularly when combined with virtualization technologies and cloud-based scientific gateways.

The project also illustrated how BOINC could support not only traditional batch scientific calculations but also machine learning workflows and large scientific cyberinfrastructures.

Although nanoHUB@Home became inactive, nanoHUB itself continues to operate as a major scientific gateway for nanotechnology research and education.

See also

• BOINC
• nanoHUB
• Volunteer computing
• Distributed computing
• Network for Computational Nanotechnology

External links

• nanoHUB@Home official website (archived) (Wayback Machine snapshot, 2024-03-30)
• nanoHUB
• BOINC