Xipeng Shen: Then and Now / 2011 Early Career

image: Winner of the 2011 Xipeng Shen Early Career Award
to see Continued

Credit: Department of Energy Science Office


The U.S. Department of Energy (DOE) Early Career Award (ECA) gave me the opportunity to advance understanding of memory performance at an extreme scale or exascale computing. This understanding is key to increasing the accuracy and scalability of many critical science simulations run on modern supercomputers.

Exascale computing systems often integrate thousands of computing units into a single chip. Each of the computer chips needs data to process. As the number of compute units keeps increasing, the total data demands increase rapidly. But the speed of moving data from memory to processors increases much slower. This growing gap fundamentally limits the currently achievable performance of exascale computing.

With the support of the ECA, I launched some research directions to reduce this gap. These directions mainly relate to reorganizing data and optimizing code on graphics processing units (GPUs), an important type of processors in exascale systems.

One of the techniques deals with irregular memory during program execution. Irregular memory accesses read or write data without patterns. They are not useful for memory systems and affect memory access speed. The new technique transforms a program so that at runtime its irregular accesses become regular. Little overhead is incurred, but data access speed increases significantly. Other techniques allow computer systems to flexibly manage the many parallel contexts on exascale systems to further speed up memory accesses.

These techniques prepared important foundations for code optimizations on exascale systems. They have influenced the development and improvements of many modern software, in the field of high performance computing and beyond. They have inspired numerous studies and received over 3,000 citations. By dramatically shortening simulation times, these techniques have helped accelerate scientific research and discovery.


Xipeng Shen is a professor of computer science at North Carolina State University.


The Early Career Research Program provides foundational financial support to early career researchers, enabling them to define and direct independent research in areas important to DOE missions. The development of outstanding scientists and research leaders is of paramount importance to the Department of Energy’s Office of Science. By investing in the next generation of researchers, the Office of Science is championing lifelong careers in discovery science.

For more information, visit Early Career Research Program.


Improved data locality of dynamic simulations for exascale calculation

Computer simulation is important for scientific research in many disciplines. Many such programs are complex and transfer a large amount of data in a dynamically changing pattern. Memory performance is critical to maximizing computing efficiency in the era of on-chip multiprocessors (CMPs) due to the growing disparity between slowly expanding memory bandwidth and rapidly increasing processor data demands.

The importance is underscored by the trend towards exascale computing, in which processors are expected to each contain hundreds or thousands of (heterogeneous) cores. Unfortunately, today’s computer systems do not support a high degree of memory transfer. This project proposes to improve the memory performance of dynamic applications by developing two new techniques specially adapted to the emerging functionalities of CMP.

The first technique is asynchronous streamlining, which analyzes an application’s memory reference patterns during runtime and regulates both control flows and memory references on the fly.

The second technique is neighborhood-aware locality optimization, which focuses on non-uniform relationships between computational elements.

This research will produce a robust tool for scientific users to improve program localization on multi- and multi-core systems, which is not possible with existing tools. Additionally, it will contribute to the advancement of computer science and foster academic research and teaching in the challenging field of scientific computing.


EZ Zhang, Y. Jiang, Z. Guo, K. Tian, ​​and X. Shen, “On-the-Fly Dynamic Irregularity Removal for GPU Computing.” Proceedings of the Sixteenth International Conference on Architectural Support for Programming Languages ​​and Operating Systemspages 369-380, (March 2011). [DOI: 10.1145/1950365.1950408]

G. Chen, B. Wu, D. Li, and X. Shen, “PORPLE: An Extensible Optimizer for Portable Data Placement on GPUs.” The 47th Annual IEEE/ACM International Symposium on Microarchitecture(December 2014). [DOI: 10.1109/MICRO.2014.20]

G. Chen, X. Shen, B. Wu, and D. Li, “Optimizing Data Placement on GPU Memory: A Portable Approach.” IEEE Transactions on Computers 66(2017). [DOI: 10.1109/TC.2016.2604372]

The DOE explains… offers simple explanations of key words and basic science concepts. It also describes how these concepts apply to work conducted by the Department of Energy’s Office of Science as they help the United States excel in research across the science spectrum. For more information on exascale computing and DOE research in this area, go to “The DOE Explains…Exascale Computing.”

Additional profiles of Early Career Research Program award recipients are available at /science/listings/early-career-program.

The Office of Science is the largest supporter of basic physical science research in the United States and works to address some of the most pressing challenges of our time. For more information, please visit www.energy.gov/science.

Warning: AAAS and EurekAlert! are not responsible for the accuracy of press releases posted on EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.

Comments are closed.