Jinpyo Kim

Email  /  Google Scholar  /  Github

profile photo
Research Papers

SnuHPL: High Performance LINPACK for Heterogeneous GPUs
Jinpyo Kim, Hyungdal Kwon, Jintaek Kang, Jihwan Park, Seungwook Lee, Jaejin Lee,
ICS '22: Proceedings of the 36th ACM International Conference on Supercomputing
Presentation (30 min)
ACM Digital Library

SnuHPL is an optimized HPL for clusters of modern heterogeneous GPUs. It generates the best HPL data distribution for a given cluster configuration by considering computing power, memory capacity, and network performance altogether. With SnuHPL, Samsung Electronic's SSC-21 supercomputer is ranked #11 in November 2021 TOP 500 List and #2 in Green 500 List.

SnuQS: Scaling Quantum Circuit Simulation using Storage Sevices
Daeyoung Park, Heehoon Kim, Jinpyo Kim, Taehyun Kim, Jaejin Lee,
ICS '22: Proceedings of the 36th ACM International Conference on Supercomputing
Presentation (30 min)
ACM Digital Library

SnuQS is a full-state quantum circuit simulation framework. It exploits storage devices, such as HDDs and NVMe SSDs, to enlarge the available main memory capacity at a small cost. SnuQS and the proposed I/O subsystem together achieves 300× cost-effectiveness over the DDR4 DRAM main-memory-only system

CyCNN: A Rotation Invariant CNN using Polar Mapping and Cylindrical Convolution Layers
Jinpyo Kim, Wookeun Jung, Hyungmo Kim, Jaejin Lee,
arXiv preprint, 2020
arXiv

CyCNN is a CNN model which exploits polar mapping of input images to convert rotation to translation. To deal with the cylindrical property of the polar coordinates, CyCNN uses cylindrical convolutional (CyConv) layers. A CyConv layer exploits the cylindrically sliding windows (CSW) mechanism that vertically extends the input-image receptive fields of boundary units in a convolutional layer.

Education

Seoul National University (Mar 2020 - Present)

Ph.D. student in Computer Science and Engineering, advised by Prof. Jaejin Lee

Seoul National University (Mar 2015 - Feb 2020)

B.S. in Computer Science and Engineering (Summa Cum Laude)

Seoul Science High School (Mar 2012 - Feb 2015)

TA Experience

Parallel Computing (Samsung Electronics System Architect, 2022 Spring)

Multicore Computing (Seoul National University, 2021 Spring)

Introduction to Quantum Computing and Information (Seoul National University, 2020 Fall)

Software Platforms for Data Science (Seoul National University, 2020 Spring)

Algorithm (Samsung SW Certificate, 2019 Spring - 2020 Summer)

Algorithm (Korea Olympiad of Informatics Winter Programming Camp, 2019 Winter)

Programming Practice (Seoul National University, 2017 Spring)

Foundation of Physics (Seoul National University, 2015 Spring - 2018 Fall)

Programming Contests

17th place, 2021 Google Hash Code world final

5th place, 2019 Samsung Collegiate Programming Cup

2nd place, 2018 ACM-ICPC Seoul Regional

3rd place, 2018 ACM-ICPC Taipei Regional

187th, 2018 Google Code Jam Round 3

3rd place, 2018 Kakao Code Festival

5th place, 2018 Samsung Collegiate Programming Cup

3rd place, 2017 ACM-ICPC Daejeon Regional

3rd place, 2017 LG Code Monster

7th place, 2016 ACM-ICPC Daejeon Regional

Scholarship

Presidential Science Scholarship (Korea Student Aid Foundation, 2015-2020)

Misc Experience

Student Volunteer of PPoPP21, PACT21, and PPoPP22.

Administrator of SNU GSDS GPU cluster (~60 nodes, ~120 GPUs)

Samsung Electronics SW Membership (2018 - 2021)

Faculty member of SAFFY (Samsung SW Academy for Youth) (June 2019 - November 2019)

Translation check of Guide to Competitive Programming: Learning and Improving Algorithms Through Contests (2019)

Samsung SDS Sgen Membership Program (2014 - 2016)

Random

Lightmap (Fast version)

Template from Jon Barron and Seohong Park's website.