The Brazos Cluster is the primary Scientific Computing resource for the Mitchell Institute and a central part of its ability to do world-leading particle physics, cosmology and astronomy. In addition to the computing power brought to the institute, CPU/Disk, the cluster is unique on campus as it is tied to the world-wide OSG Grid Computing Infrastructure used heavily by the CMS and CDMS groups. The loss of this invaluable resource would both renege on our extensive commitments to our experiments, and likely make us non-competitive for a number of years as we struggled to rebuild. The Brazos Cluster is a Tier 3 grid computing center for the CMS experiment, and a Tier 2 computing center for the CDMS experiment. This has directly allowed our groups to play leadership roles in the experiments, as well as provided us with a world-wide competitive advantage in the analysis of the data, which directly affects our ability to do competitive science (and for us to receive funding for doing it). Many students have gotten their PhD using the cluster. It is also special as it is an excellent example of joint effort between the cluster admins and the scientists in the group to support the software and interfaces that keep the jobs running for the CDMS, CMS and Phenomenology groups that use it so heavily. As such, it is excellent training for our students and postdocs in big data/big computing. It has even landed a number of students permanent positions in big data/big computing at the national labs because of their training. In sum, the Brazos cluster is a vital resource, one that makes the Institute a great place to get science done, and a great place to get cutting edge training in science and big data. The future is equally clear as the trend towards big data and big computing accelerates. This is most evident in the expected increases running with the recent turn-on of the LHC and the big simulations expected for CDMS during the design/construction phase. The cluster is well leveraged as the amount of resources from the Mitchell Institute, as well as the Department of Energy, the Norman Hackerman ARP program, and the American Recovery and Reinvestment Act.
Although the Academy’s support for the application of cyberinfrastructure to physics and applied mathematics is not surprising, our work with A&M’s Digital Humanities group might be. Dr. Laura Mandell, director of the IDHMC, is also an active researcher with the IDHMC, with a particular interest in using digital technologies to develop and apply high-quality text resources for use in research on literature and culture. Specifically, she and her group work with the Academy to organize a set of images of more than 40,000,000 pages of English language printed literature between 1470 and 1800 and then to use advanced optical character recognition (OCR) software to convert those images to text. Key to this work is Trey’s Dockendorf’s expertise at setting up, operating, and maintaining computer systems to store and curate these data sets, work with IDHMC experts on web-based (including crowd-sourcing) applications, and enable high-speed access to the data sets from the Brazos cluster. Based on this work IDHMC researchers are beginning to use Brazos and modern OCR software to translate millions of page images to page text. When this work is complete, Dr. Mandell intends to have the entire corpus of early English language printed literature (including legal filings and even newspapers from Ireland and India) available as text for scholarly study. This project reflects both the power and flexibility of Brazos and also the skill of Academy personnel in addressing a wide variety of scholarly research needs.
The Academy is one of several A&M centers collaborating through the IAMCS. As a contributor to the IAMCS, it has taken a lead role specifically in cyberinfrastructure issues, including leveraging its experience with the Brazos Cluster to help create an advanced IAMCS High-Performance Computer. Of equal importance, it is helping to integrate key cyberinfrastructure resources, including Brazos, the Immersive Visualization Center, advanced mass data stores, and advanced networks such as LEARN and Internet2. This infrastructure effort within the IAMCS plays a significant role in the IAMCS’s award from the King Abdullah University for Science and Technology (KAUST). Further, it is helping integrate these campus resources, as appropriate, with those of KAUST. The result will be a state-of-the-art, multidisciplinary computational research base with a robust infrastructure that enables large-scale scientific computation research, advanced multidisciplinary learning and education, and a large number of applications in diverse areas of science and engineering. The IAMCS Hurr high-performance computer was operated and supported IAMCS research, and a high-speed 100-TByte parallel file system was continued in operations. Hurr is similar to Brazos in many respects, but (a) has 128 compute nodes and (b) uses 16-Gb/s Infiniband interconnect fabric instead of the gigabit Ethernet fabric used on Brazos. As mentioned earlier, there are significant synergies between Hurr and the high-throughput Brazos cluster.
The ISC is a multidisciplinary research center devoted to designing, analyzing, and implementing innovative computational tools that advance scientific engineering research and education. ISC researchers include internationally recognized Texas A&M faculty members devoted to collaborating on major national and global research efforts with other universities, industrial partners, and the government. The ISC also serves as an excellent training ground for students, both graduate and undergraduate, and postdoctoral scholars in a variety of academic disciplines within scientific computing technologies.
The Laboratory for Molecular Simulation (LMS) brings molecular modeling and computational chemistry closer to the experimental scientist by offering training to both new and advanced users. Advanced modeling software is available for researchers at Texas A&M University to perform quantum calculations on small molecular or solid systems and molecular mechanics/dynamics modeling for large systems such as proteins, DNA, nanomolecules, polymers, solids, and liquids. The LMS also provides support for faculty that wish to incorporate molecular modeling in their course material.
Despite enormous investment in the development of defensive cyber capabilities we, as a nation, continue to lose ground against determined adversaries in the conflict for control in cyberspace. The Texas A&M Cybersecurity Center is dedicated to combating adversaries who desire to harm our citizens, our government or our industry through cyber-attacks. The Center seeks to advance the collective cybersecurity knowledge, capabilities and practices through ground-breaking research, novel and innovative cybersecurity education, and mutually beneficial academic, governmental and commercial partnerships. Working with researchers, faculty and industry leaders, the Center stands committed to make outsized contributions to social good through the development of transformational cybersecurity capabilities. TAMC2 was previously known as the Center for Information Assurance and Security (CIAS).