ARC released a new cluster named Huckleberry in late 2017. The Huckleberry system, accessed at huckleberry1.arc.vt.edu, was installed with deep learning applications in mind. To this end, it consists of 14 IBM “Minsky” S822LC nodes and NVIDIA’s proprietary NVLink interconnect network. This system enables highly parallel and highly distributed workloads. IBM unveiled its deep learning AI toolkit called PowerAI alongside the launch of Minsky nodes that leverage CPUs linked to Power CPUs with NVLink making it possible to have high speed high performance computing. PowerAI is available under
/opt/DL in Huckleberry.
Each compute node on Huckleberry (i.e. IBM “Minsky” nodes) consists of :
- Two IBM Power8 with 10 cores, 8 threads per core and memory bandwidth 115gb/s per socket
- Four NVIDIA P100 GPUs advertised to have 21 teraFLOPS of 16-bit floating-point performance ideal for deep learning applications deliver high performance, massive parallelism
- NVIDIA’s NVLink technology which provides high bandwidth data transfers between CPUs and GPUs; an improvement over PCI-Express
- Mellanox EDR Infiniband (100 GB/s) interconnect used to connect compute nodes
The PowerAI toolkit contains Caffe, TensorFlow etc. which are optimized for the Power servers. IBM provides support for it as well.
While the rest of the clusters make use of the PBS batch systems, Huckleberry makes use of the Slurm batch system using the command
Individuals may request a Huckleberry account. Instructors can get set up class accounts for Huckleberry as well.
ARC is happy to announce the addition of 39 new GPU nodes to the NewRiver cluster. Each of these nodes is equipped with two Intel Xeon E5-2680v4 (Broadwell) 2.4GHz GPU (28 cores/node in all), 512 GB memory, and two NVIDIA P100 GPUs. Each GPU is capable of up to 4.7 TeraFLOPS of double-precision performance, so including CPU and GPU these nodes add over 400 TFLOPS of peak double-precision throughput to ARC’s resources.
Continue reading P100 GPU Nodes added to NewRiver
HPC Day Agenda
We are looking forward to seeing you at our Annual HPC Day event March 24 from 10am-5pm!
The event includes: a keynote by James Ahrens from LANL, a machine learning workshop, and faculty and student presentations.
Keynote: “Supercharging the Scientific Process Via Data Science at Scale”
Dr. James Ahrens is a senior research scientist at the Los Alamos National Laboratory (LANL). He is the founder and design lead of ParaView, a widely adopted visualization and data analysis package for large-scale scientific simulation data ( http://paraview.org). ParaView has had an extremely positive impact on the large-scale data analytic capabilities available to simulation scientists around the world. Dr. Ahrens graduated in 1989 with a B.S. in computer science from the University of Massachusetts and in 1996 with a Ph.D. in computer science from the University of Washington. At LANL, he is part of a data science team of twenty staff, postdocs and students. He is also a national leader of programmatic initiatives important to the United States Department of Energy’s National Nuclear Security Administration and Office of Science. Dr. Ahrens is the Data Analysis and Visualization lead for the U.S. Exascale Computing Project and the general chair for this year’s IEEE Scientific Visualization conference to be held in Phoenix, AZ in early October.
ARC member Nicholas Polys helped organize a session at the CHCI Workshop Technology on the Trail on March 2-3. The session “From Experience to Abstraction and Back Again” discussed the challenges and opportunities for citizen science, especially the use of uncertain data to build new scientific models. The event was covered with an article in the Roanoke Times!
Featuring sessions on big data workflows, data visualization, data publishing, and reproducible research practices, the 2017 Big Data Science Workshop will also incorporate a brainstorming/strategy session aimed at improving research workflows, a networking breakfast, and lightning talks.
ARC’s Nicholas Polys and Brian Marshall each presented. The event flyer is here:
Big Data Science Workshop
The Visionarium got a spot in the 2016 Hokie Halftime commercial, which aired during Virginia Tech’s first football game against Liberty University. Check us out and see what we have been up to recently! Go Hokies! https://www.youtube.com/watch?v=Jl9iL2a-pmw
Our last halftime spot was 2012 : https://www.youtube.com/watch?v=p8nER5wb6cA
ARC is happy to announce the release of a new cluster, named DragonsTooth, available at
dragonstooth1.arc.vt.edu. DragonsTooth is made up of 48 nodes, each equipped with:
- 2 x Intel Xeon E5-2680v3 (Haswell) 2.5 GHz 12-core CPU (same CPU as NewRiver)
- 256 GB 2133 MHz DDR4 memory for large-memory problems
- 4 x 480 GB SSD Hard Drives for fast local I/O ($TMPDIR)
- 806 GFlops/s theoretical double-precision peak
Continue reading New ARC Cluster: DragonsTooth
Dr. Srijith Rajamohan presented an Introduction to Python Pandas for Data Analytics tutorial. Pandas is a high-level open-source library that provides data analysis tools for Python. The audience was also introduced to relevant packages such as Numpy for fast numeric computation and Matplotlib/Bokeh for visualization to supplement the data analysis process. The slides for this tutorial can be found here.
Visualization GRA and Doctoral Candidate Ayat Mohammed presented a visualization showcase titled ‘Insights into Alzheimer’s Disease: Molecular Dynamics (MD) Simulations of Peptide-Membrane Interactions’ at XSEDE16, Miami. Also from ARC, Alana Romanella chaired the session on Workforce Development and Diversity.
Dr. Srijith Rajamohan and Dr. Nicholas Polys in ARC and Assistant Professor, Vivica Kraak in the Department of Human Nutrition, Foods, and Exercise are collaborating on a research project to map the world of celebrity endorsement of food and beverage brands, products and groups in the United States. HNFE doctoral student, Mi Zhou, is part of the research team with ARC MS student, Faiz Abidi, to build, analyze and visually display in 2D and 3D a database of more than 550 unique celebrities used to market food, beverage and restaurant products to children, teens and adults. The ARC team had helped build an open-source analytics and visualization engine to help address these needs.
The results of this project will be used to inform the policies and actions of diverse stakeholders including industry, government and public health groups to use celebrity endorsement, along with other integrated marketing communications, to promote healthy nutrient-profile products and behaviors that support healthy food environments for American children, adolescents and their parents. Prof. Kraak and her work was recently featured on the VT news which can be found here: