Over the past year, Prof. Chandrasekaran and Robert Searles have teamed up with researchers Wayne Joubert and Oscar Hernandez of Oak Ridge National Lab in an effort to accelerate a miniapp called Minisweep. Minisweep is representative of the main computational kernel of a production Sn radiation transport application called Denovo, which is used for nuclear reactor neutronics modeling. This code is instrumental for nuclear scientists because modeling neutron flow helps them avoid a nuclear meltdown, and it provides them with information that is essential in designing an effective shielding system around the reactor to protect workers from excessive radiation exposure.
Prof. Chandrasekaran gave an HPC featured talk at the yearly GPU Technology Conference (GTC), San Jose, CA, 2018 on Adapting Minisweep, a Proxy Application, on Heterogeneous Systems Using OpenACC Directives based on work done by her PhD student Robert Searles. This is work in collaboration with Dr. Oscar Hernandez and Dr. Wayne Joubert at the Oak Ridge National Laboratory (ORNL). The talk highlighted programming challenges and its corresponding solutions using OpenACC, for porting a wavefront based miniapplication for Denovo, which is a production code for nuclear reactor modeling.
CRPL member Arnov Sinha joins Numeca International as an intern to port Computational Fluid Dynamics code to large scale supercomputers using high-level directive-based programming model, OpenACC. Arnov’s defended his Master’s thesis on’“High Performance Sparse Fast Fourier Transform using OpenACC.”
CRPL member Kshitij Srivastava leaves for Oak Ridge National Laboratory to work on cancer data and deploy deep learning strategies on large scale supercomputers. Kshitij Srivastava graduated with his Masters degrees from Electrical & Computer Engineering (Spring 2017). He can be reached on LinkedIn or on Twitter“
On June 5-9, 2017, CRPL’s Robert Searles, Kyle Friedline, Sanhu Li, and José Monsalve Díaz traveled to Brookhaven National Laboratory (BNL) to participate in the GPU Hackathon hosted there. BNL hosted this event in conjunction with NVIDIA, The Portland Group (PGI), and Oak Ridge National Laboratory (ORNL). Robbie and Kyle acted as mentors for their teams, providing insight using their extensive GPU programming experience. Robbie served as the mentor for Team Uduh: a combination of members from University of Delaware and University of Houston.
Our two teams presented at GTC 2017. Sanhu Li, Sunita Chandrasekaran “S7341: Using OpenACC for NGS Techniques to create a portable and easy-to-use code base” May 9, 15:00-15:25, Room 210C slides Arnov Sinha, Sunita Chandrasekran “S7478: Using OpenACC to Parallelize Irregular Algorithms on GPUs”, May 10, 15:30 - 15:55, Marriott BallRoom 3 slides
CRPL members Arnov Sinha and Sergio Pino graduated with their Masters degrees in Computer & Information Sciences in Summer 2017. Arnov’s defended his Master’s thesis on’“High Performance Sparse Fast Fourier Transform using OpenACC”. Kshitij Srivastava graduated with his Masters degrees from Electrical & Computer Engineering (Spring 2017).
Our PhD student Sergio Pino’s paper on “Exploring translation of OpenMP to OpenACC 2.5: Lessons learned” has been accepted to the 7th AsHES co-located IPDPS 2017 - Sergio will be giving his talk on May 29 at AsHES workshop
The next GPU Hackathon called “Brookathon” is coming to the Town of Brookhaven! This Hackathon will take place at Brookhaven National Laboratory on June 5 to 9, 2017, and is jointly organized by Brookhaven National Laboratory, University of Delaware, Oak Ridge National Laboratory, and Stony Brook University.