Current Research Projects

Volunteer Computing:
Docking@Home: Dynamically adaptive protein-ligand docking system using volunteer computing
Docking@Home Screensaver: Docking@Home screensaver and graphical application for Volunteer Computing Projects studying molecular systems.
EmBOINC: An Emulator for BOINC Applications

GPU Programming and Scientific Applications:
FEN ZI (yun dong de FEN ZI = Moving MOLECULES) a CUDA code that enables large-scale, GPU-based MD simulations using the CHARMM force field. In FEN ZI, the complete MD simulation is performed on GPUs.
DACAPO GPUs - Developing Applications for Computer Architectures based on Parallel Organizations of GPUs: Use the computing power of parallel GPUs for scientific applications, e.g., Molecular Dynamics and Monte Carlo simulations (web-page under construction)

Grid Computing:
jTopaz: A friendly, open-source GridFTP protocol extension to the Firefox browser for accessing grid repositories

Bioinformatics:
RNAVLab: An open-source user-friendly virtual Laboratory for RNA secondary structure predictions
Pseudobase++: A searchable up-to-date database of the PseudoBase pseudoknots wrapped by a versatile, user-friendly interface providing scientists with a powerful engine to access, search, select, and sort data based on different fine-grained criteria

Our Research Equipment

Our laboratory is equipped for studying scientific applications on multi-core and GPU platforms. We have access to several clusters at the University of Delaware and high-end workstations in our lab.

Dr. Taufer was awarded an NSF Major Research Instrumentation (MRI) award (CHE-0922657). The NSF award supported the acquisition of a hybrid-computing cluster in 2011 with GPU-accelerated computing nodes. The cluster includes 48 dual six-core compute nodes (576 cores), 96 Fermi S2070 GPU systems. Two front-end nodes for compilation and workload management as well as a 20 TByte storage system are connected to the compute nodes. The cluster utilizes both an Infiniband fabric and a Gigabit Ethernet interconnect. The GPU-enabled capacity of the cluster supports implementation and testing of HPC research involving multi-threading GPU programming in scientific computing. The cluster also supports a large number of theoretical and experimental researchers at UD to study a number of problems in the chemical sciences. Our equipment resources include a small high-end cluster comprising of:

  1. A dual quad-core compute nodes (48 cores) and 3 Tesla S1060 GPU systems with 4 GPUs per Tesla, 12 GPU total. Each Tesla system is connected to two compute nodes.
  2. A dual quad-core compute nodes (16 cores) each hosting 4 Fermi C2050 GPUs, 8 GPU total.
A front-end node is connected to the compute nodes and is used for compilation and job submissions. A high-speed DDR Infiniband interconnect for application and I/O traffic and a Gigabit Ethernet interconnect for management traffic connects the compute and front-end nodes. In our lab we work with high-end dual quad-core workstations, each hosting two GPUs, either NVIDIA GeForce GTX480, or GTX 580 (Fermi).

Software For Download

The D@H screensaver source is now availabe in Google Code. Download the code here.

Past Projects

SHiPPER: Spreading High-Performance computing Participation in undergraduate Education and Research (Old web-page at UTEP - page no longer maintained by M. Taufer)
MSM Consortium - Working Group 5: Working Group targeting topics in high-performance computing, computational issues, and algorithms
Predictor@home: Protein structure predictions using volunteer computing (TSRI site)