TAESUNG LEE

Ph.D. (2017. 2 ~) @ POSTECH, South Korea, Dept. of Computer Science Engineering

B.S. (2012. 2 ~ 2017. 2) @ POSTECH, South Korea, Dept. of Electronic and Electrical Engineering

Ko, S., Lee, T., Hong, K., Lee, W., Seo, I., Seo, J., and Han, W., "iTurboGraph: Scaling and Automating Incremental Graph Analytics," In Proc., 47th Int'l Conf. on Management of Data, ACM SIGMOD, Xian, Shaanxi, China, June 2021.

Participate in the project 'Development of a distributed graph DBMS for intelligent processing of big graphs' as a main PIC (2021 - 2024)
Ph.D. Student, POSTECH (IITP)

Participate in the project 'A Trillion-Scale Graph Processing Engine for Incremental and Temporal Graph Analytics' (2018 - 2021)
Ph.D. Student, POSTECH (삼성 미래기술 육성 사업)

Scalable and Automated Incremental Graph Analytics System (2018 - 2020)
Ph.D. Student, POSTECH

• Design and implement a dynamic graph store to support updates for vertex attribute values and edge connections. To avoid costly in-place updates, we propose a delta-based maintenance strategy.

• Implement and optimize the graph processing engine module of iTurboGraph, resulting in up to 43.9X speedups by the incremental computations.

• Published at SIGMOD 2021.

Participate in the project 'Trillion-scale Big Graph Processing Engine' (2017 - 2018)
Ph.D. Student, POSTECH (삼성 미래기술 육성 사업)

Scalable and Fast Graph Analytics System (2017 - 2017)
Ph.D. Student, POSTECH

• Implement a Balanced Buffer-Aware Partitioning (BBP) algorithm that achieves 1) balancing the workloads across machines to maximize the overall efficiency of graph processing, 2) balancing the number of vertices across machines so that each machine requires the same memory budget, 3) using the range partitioning for fast accessing of vertices.

• Design and implement a network I/O layer of TurboGraph++.

• Published at SIGMOD 2018 ("TurboGraph++: A Scalable and Fast Graph Analytics System).

Graph Database Management System

Distributed Graph Processing Systems

Temporal Graph