Yifeng Cui
University of California-San Diego, CA·Computer Science
Earthquake simulations on supercomputers are becoming key tools for guiding infrastructure and policy decisions. These codes split their work across many computer nodes that must constantly exchange data.
- Fields
Computer Science
Earth & Environmental Science
Engineering
- Topics
high-performance computing
supercomputing
Message Passing Interface
- Grants
- 1 known grant, through Aug 2027
What they work on
- Couple the MVAPICH MPI library and the TAU profiling system through the MPI_T interface so communication settings can be monitored and tuned while scientific codes run.
- Speed up AWP-ODC 3D earthquake simulations using load-aware asynchronous and partitioned point-to-point communication designs from the MPI 4.0 standard.
- Improve heFFTe Fast Fourier Transform performance with application-aware neighborhood collectives, adaptive persistent collective communication and coordinated GPU communication.
More about the research
That communication often limits speed, and tuning it at very large scale is hard. New computer hardware with many-core processors and GPUs (graphics processors), along with new features in the MPI 4.0 standard (the common rulebook for passing messages between nodes), creates chances to redesign how applications communicate.
Why it matters
Faster, better-tuned earthquake simulations could improve earthquake forecasting and our understanding of seismic events. That could help reduce risks to cities and the environment. The improved communication and profiling tools could also speed up other scientific codes that run on next-generation supercomputers, including those that rely on Fast Fourier Transforms.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Neranjan Edirisinghe Pathirannehe
Georgia State University, GA·Computer Science
Deploy a campus computing cluster with 12 NVIDIA H100 GPUs, BlueField data processing units, 682 CPU cores and 228 TB of storage to support 11 research projects spanning wildfire management, solar prediction, neuroscience, finance and policy.
Chengming Zhang
University of Houston, TX·Computer Science
Adapt GPU error-bounded lossy compressors such as cuSZ to real scientific workflows, supporting diverse data formats and compression quality targets, with quality checked by tools like QCAT.
Ryan Johnson
University of South Dakota, SD·Computer Science
Add computing capacity, including GPU-accelerated nodes, to the Lawrence high performance computing cluster to speed research in clean energy, sustainability, PTSD and climate change across South Dakota.
The school
University of California-San Diego: 84 labs on PI Pages, $1.9B a year on research, 637 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.