The Monte Carlo (MC) method has been widely regarded as the gold-standard for modeling light propagation inside complex random media, such as human tissues. It, however, suffers from low computational efficiency because a large number of photons have to be simulated to achieve desired solution quality with low stochastic noise. Sequential MC simulation implementations require extensive computation and long run-time, typically ranging from several hours to several days.

In this work, we use Open Computing Language (OpenCL) to leverage the computing power of GPUs. We analyze the characteristics of MCXCL kernel and develop several optimization schemes to improve the execution efficiency such as branch divergence. Previous static workload distribution scheme at the thread-level has been improved by the dynamic scheme at the work-group level. The workload balancing issue, when marshaling simulation among various GPU devices, is addressed by the proposed throughput model.

Project Github PPT

Cellulose is one of the most promising energy resources that is waiting to be tapped. Harvesting energy from cellulose requires decoding its atomic structure. Some structural information can be exposed by modeling data produced by X-ray scattering.

In this work, we accelerate a molecular scattering algorithm by leveraging a GPU cluster. The optimization approach considers memory utilization, math intrinsics, concurrent kernel execution and workload partitioning, using CUDA, OpenMP and MPI.

Leveraging accelerators hosted on a cluster, we have reduced days/weeks of intensive simulation to parallel execution of just a few minutes/seconds. Our GPU-integrated cluster solution can potentially support concurrent modeling of hundreds of cellulose fibril structures, opening up new avenues for energy research.


The National Institute of Environmental Health Sciences supported Super-Fund Research Program, Puerto Rico Test-site for Exploring Contamination Threats (PROTECT) is a large multi-institution program that investigates the role environmental pollutants play in preterm birth, a major factor causing over one-third of infant deaths in USA.

I am the database administrator (2014-206) for the Data Management and Modeling Core (Core D). I am responsible for the following tasks.

  • Provide domain specific upload formats and on-demand data exports.
  • Maintain data integrity and featured dependencies.
  • Support multi-dimensional data mining.
  • Offer customized analytical reports for distributed researchers.
  • Safe dissemination of sensitive environmental, health, and biological data. (EQUIS Enterprise)

In collaboration with Project 3 and Xiangyu Li, we published a conference paper and a journal, related to the big data management and analysis.

Read More ...