Research Innovation Drives an Industry-Leading Computational Geometry Engine in High Speed

Posted: December 1, 2016

The polygon overlay is a complex and time-consuming process to superimpose multiple geographic layers and their attributes to produce a new polygon layer. This process has become increasingly massive in the big data era from various applications, such as graphical information systems, electronic design automation, computer vision, image processing, and motion planning solutions for robotics.  The industry demands fast and efficient solutions for daily production tasks of spatial data analytics in many areas. A research innovation led by a group of Computer Science and Engineering researchers at Ohio State has timely responded to this need.

Dr. Akihiro Asahara, the CEO of Fixstars Solutions Inc. recently sent Dr. Kaibo Wang (CSE Ph.D’15) an acknowledgement letter to inform him that Fixstars has effectively developed the Geometric Performance Primitives (GPP) Library, an industry-leading and high speed computational geometry engine, based on Wang’s work published in VLDB 2012. Dr. Asahara states, “Specifically, the PixelBox algorithm of yours lays a scientific foundation for massive polygon overlay operations, which enables us to achieve a huge performance advantage (up to 25 times faster) over other similar industry products.” GPP has also been included in the GPU-Accelerated libraries of the NVDIA Company.

PixelBox is a fast parallel algorithm for massive polygon overlay operations, which is implemented in a hybrid systems of both GPUs and multicore processors, and tested by pathology image analysis workloads from hospitals. This work entitled “Accelerating Pathology Image Data Cross-Comparison on CPU-GPU hybrid Systems” was presented in the 38th International Conference on Very Large Databases in August 2012 in Istanbul, Turkey, and was published in the Proceedings of the VLDB Endowment, No. 5, No. 11 in 2012. The authors of the paper are Kaibo Wang, Yin Huai, Rubao Lee, Fusheng Wang, Xiaodong Zhang, and Joel H. Saltz.

Both Kaibo Wang and Yin Huai received their Ph.Ds. in Computer Science and Engineering at The Ohio State University in 2015 under the supervision of Professor Xiaodong Zhang. They now work at Google and Databricks, respectively. As students, each received the Department of Computer Science and Engineering Graduate Research Awards.

The other authors are homed elsewhere. Rubao Lee is a Research Scientist in OSU-CSE. When the paper was published, Drs. Fusheng Wang and Joel Saltz were on Faculty in the Bioinformatics Department at Emory University, but are now faculty members at SUNY Stoney Brook.

“I am very pleased to see another basic research work of ours directly impacts on production systems, which is a high recognition to the value of our research efforts”, says Xiaodong Zhang, the Robert M. Chritchfield Professor in Engineering and Chair of Computer Science and Engineering at Ohio State. Several published research results in computer systems and data management from his group have been widely adopted in production systems of both hardware and software.

From left to right: Yin Huai, Xiaodong Zhang, Kaibo Wang, and Rubao Lee.
Category: Department