We are pleased to welcome Dr. Thomas Furlani to the UBIT blog. Dr. Furlani has been involved in high-performance scientific computing and data visualization at UB for nearly 20 years.
The past decade has witnessed a revolution in the application of information technology to scientific research, fueled in part by the explosion of data from high-throughput experimental devices.
For example, genome sequencers are now available that can produce more than 4 terabytes of data per run. This is only the beginning, as companies are developing sequencer advances with the potential of producing upwards of a billion bases of sequence every five minutes. Putting this in perspective, this translates to sequencing the human genome in less than 4 hours (sequencing the human genome the first time took more than 10 years!).
The storage, management, and analysis of today’s enormous data flows requires substantial computing power and immense mass-storage capacity. The same is true in many other areas of science and engineering. Accordingly, access to advanced computational resources by researchers from a wide variety of fields (from traditional sciences and engineering to medical research and the humanities/social sciences) is no longer a luxury, but rather a necessity if they hope to stay at the forefront of their respective field, advance knowledge and improve the quality of life.
Fortunately, through the Center for Computational Research (CCR), UB is well-positioned to meet the computing and data analytics needs of its researchers. CCR’s extensive computing facilities currently include a Linux cluster with more than 8,000 processor cores, and about 600 Terabytes of high performance storage, which will soon undergo an upgrade to more than 3,000 Terabytes (3 Petabytes).
In total, CCR currently has more than 100 Tflops of peak performance compute capacity and, by way of comparison, its computers can do in one day what a high-end PC would take more than 25 years to do. CCR’s computers run 24 hours a day, 7 days a week. Last year alone, more than 2 million jobs ran on CCR resources by UB faculty and their students.
This number will grow dramatically over the next several years as CCR helps support the newly formed Buffalo Institute for Genomics (BIG) and Data Analytics, which was created in 2014 to provide informatics and biomedical expertise to advance genomic medicine. This is part of a $100 million collaboration between the University at Buffalo and the NY Genome Center in New York City.
UB was selected as co-lead for the effort because of its expertise in high performance computing, combined with recognized national leadership in genomics and analysis of patient data. To help support BIG and the NY Genome Center, CCR will substantially expand its high performance computing and storage infrastructure.