Modern scientific research often generates huge, diverse amounts of information. The ability to analyze extremely large, complex amounts of data can reveal patterns, trends, and associations that are vitally important to advancing research and finding breakthroughs.
To meet the demands of ever-growing datasets that require computationally intense analyses, the Center for Information Technology (CIT) developed Biowulf, NIH’s supercomputer. Biowulf is a general-purpose, high-performance computing resource capable of analyzing petabytes of data. Biowulf is used across all NIH institutes and centers in a variety of domains, including genomics, structural biology, image analysis, computational chemistry, artificial intelligence, machine learning, and more.
Biowulf Enables the First Complete, Gapless Human Genome Sequence
One of Biowulf’s most important contributions happened in 2022, when the Telomere-to-Telomere Consortium, a global team of scientists led by the National Human Genome Research Institute, reported they had published the first complete, gapless human genome sequence. It was a landmark achievement, and Biowulf’s large scale computing and storage resources helped by enabling geneticists to sequence and study areas of the human chromosome that contain highly repetitive DNA that had long remained a mystery.
Biowulf Keeps Up with Growing Demand
To ensure Biowulf kept up with research community needs, CIT significantly enhanced the system starting in 2015, resulting in a 500% increase in computing capacity, a 1,400% increase in storage capacity, and support for over 1,000 shared biomedical applications.
Biowulf is the world’s most powerful supercomputer solely dedicated to advancing biomedical research and is used by 75% of NIH Principal Investigators across NIH represented by more than 2,400 active users. Biowulf also attracts top data science and research talent to the NIH intramural research program.
In FY2024, NIH researchers ran computational analyses requiring more than a billion core-hours and approximately 6 million GPU-hours on Biowulf. Biowulf has also been cited by nearly 6,000 peer-reviewed publications since its early development in the late 1990’s. In fact, in 2024, about 10.8% of all published papers at NIH acknowledged the use of Biowulf.
Teaching Biowulf to NIH
CIT’s High-Performance Computing team not only supports Biowulf’s hardware and software, but they also help researchers get the most out of the system by providing classes, seminars, and walk-in consultations. The team has the technical and scientific know-how to handle a variety of concerns, from scripting problems to node allocation to helping develop strategies for particular projects.
Biowulf Powers Pandemic Research
During the pandemic, CIT prioritized Covid-19 research on Biowulf. NIH researchers used Biowulf to examine the structure of the virus, study genetic variations, and investigate pathogenic characteristics of the virus. Covid-19 projects running on Biowulf consumed over 87 million CPU hours and resulted in more than 50 peer-reviewed scientific publications.
Learn More about Biowulf
CIT maintains several hundred scientific programs, packages and databases for Biowulf users. Check out the full list of system-installed software along with additional information on the Biowulf website.