A seasoned senior software engineer with over a decade of experience crafting open-source software (OSS) for HPC systems, hailing from a global background of a life lived in South Korea, the USA, Germany, and Spain.
- 12 years of experience creating OSS for HPC systems.
- Tech lead at DoE project ECP ADIOS2, a widely used software among the world's fastest supercomputers.
- Enthusiastic about C/C++ and System Programming.
- Experienced OSS contributor and package maintainer (Spack, OpenSUSE, Pypi and Conda).
Senior R&D Engineer
Kitware, New York• February, 2020 — present
Seasoned senior software engineer with over a decade of experience crafting open-source software (OSS) for HPC systems.
- ECP ADIOS2: Tech lead, Release engineer, Core developer. In charge of deployments in current fastest supercomputers (Frontier/Summit/Crusher).
- ParaView: Core Developer, Release Engineer.
- VTKm: Core developer, Release Engineer. In charge of deployments in current world fastest supercomputers.
- VTK: Core Developer, Release Engineer.
BeeGFS Core developer
ThinkParQ, Germany• April, 2019 — January 2020
As a core member of BeeGFS developing team I implemented features in both user-space and kernel-space. I also ported its in-house RDMA network library to onmi-path among many other things.
Research Assistant
Data Intensive Computing Lab, South Korea • Feb, 2017 — Feb, 2019
As a RA during my graduate school, I continued the projects that I had been working earlier during my undergraduate in the same lab (EclipseMR and Velox, see below). During this years I attended or published in the following conferences:
- SC'2018 (Dallas, USA), participant.
- IEEE Cluster 2017 (Honolulu, USA), main author of EclipseMR
- IEEE/ACM CCGRID 2017 (Madrid, ES), participant.
- ACM SOSP 2017 (Shanghai, CN), participant
Big Data Systems Architect
Fraunhofer IOSB-INA, Germany • Dec, 2017 — Feb, 2018
As a member of the Big Data department, my role in this position was to provide big data solution for industry 4.0 challenges. During my stay at Fraunhofer IOSB-INA I designed and developed the functional prototype of what it will later might become an universal plataform for big data technologies at the industrial automation area.
Undergraduate Research Assistant
Data Intensive Computing Lab, South Korea • June, 2012 — Feb 2017
I have been working as an undergraduate lab intern in DICL lab since few month after my arrival at Korea. My stay of four years in this lab has gave me an unique experience in the computer sciences research challenging me to design and implement distributed frameworks which aim to compete current ones. Also, as a lab member I was enabled to attend several seminars from many great scientist from all the world.
- Velox architect and main developer.
- Simring prototype architect and main developer (experimental distributed cache).
- Teacher Assistant and official tutor for Engineering programming courses.
Participant startup, Berkeley Method of Entrepreneurship bootcamp (BMOE'16)
University of California, Berkeley • Dec, 2015 — Feb, 2016
Ulsan National Institute of Sciences and Technology, UNIST
Master degree of Computer Sciences Engineering
Ulsan National Institute of Sciences and Technology, UNIST
Double undergraduate degree of Computer sciences engineering and entrepreurship
Universidad de Cadiz, Spain
Computer Sciences Engineering Bachelor
Projects and Publications
ECP ADIOS2
Tech Lead, Kitware • 2021 — present
Department of Energy (DoE) project, a widely used software among the world's fastest supercomputers. Responsible for deployments in current fastest supercomputers (Frontier/Summit/Crusher).
Paraview
Core Developer, Kitware • 2020 — present
World leading OSS computer visualization suite5.
VTKm
Core Developer, Kitware • 2020 — present
HPC port of the world leading OSS visualization toolkit (VTK)6.
BeeGFS
Core Developer, ThinkParQ • 2019 — 2020
World leading performance oriented Distributed File System4.
FactoryStreaming
Architect, Fraunhofer IOSB-INA • 2017 — 2018
FactoryStreaming is big data solution to analize, store and visualize real time data from Smart factories. It uses technologies such as stream processing, data pipelining, time series databases and industrial protocols
EclipseMR: Distributed and Parallel Task Processing with Consistent Hashing
Main author, IEEE cluster 2017
In this work I presented a novel MapReduce framework prototype that efficiently utilizes a large distributed memory in cluster environments2.
Velox big data framework
Creator and primary maintainer• 2015 — 2019
Big data processing and storage framework based on EclipseMR (see below) which is design on the foundations of a Distributed File system that can intelligently adapt to changing workload in distributed systems, efficiently utilizing all the resources of the system3.
Sunny Side Up Systems
Co-founder and CTO • 2015 — 2016, Berkeley, CA
Attempt to commercialize and develop EclipseMR (see below). While the startup was funded and based in Korea, for a large part of its timeline we moved to Berkeley, CA, as part of a program provided by our university to learn and get assistance from famous business incubators and from other similar startup market leaders.
Asian supercomputing challenge (ASC'14)
Finalist, Sun-yat sen University, Guangzhou, China • 2014
Only foreigner in the South Korean team of 5 members in ASC'14. During this competition, our team moved to Guangzhou in China where inside the Tianhe-2 building and within a power budget of 3000 kwh, we built and optimized our HPC cluster to perform the given software as most efficiently as possible. 16 countries participated in this competition. Most of our work relied in the kernel level settings/hacks of our CentOS, building optimization and tuning the MPI framework.
International Supercomputing Conference (HPCAC-ISC'14)
Finalist, Leipzig, Germany, • 2014
Three months after China, our team participated this new challenge. The competition was just another event inside the International Supercomputing Congress. Again the constraint of 3000 kwh was required, however this time our team brought the HPC cluster previously assembled in our university at Korea1.
GUKKET (Startup Project)
Co-founder, South korea, • 2014
As required in my undergraduate program, I was assigned to develop a startup with a small team, our idea was an online second hand market platform for students living in dormitories. We successfully implemented the system and ran it in the local community.
UNIST Outstanding Contribution Award
Outstanding Achievement • 2012 — 2014
The highest undergraduate student honor award at UNIST.
UNIST international Student Organization (UISO)
Cabinet member, • 2012 — 2013
As a cabinet member, our mission in the direction of UISO was leading and implementing improvements for the stay of the foreigners student at UNIST (at that time more than 150 students). This experience strongly helped me to build my leadership and initiative.
1: ISC14 UNIST TEAM
2. V. A. B. Sanchez et al., "EclipseMR: Distributed and Parallel Task Processing with Consistent Hashing," 2017 IEEE International Conference on Cluster Computing (CLUSTER), Honolulu, HI, 2017, pp. 322-332, doi: 10.1109/CLUSTER.2017.12.
3: VeloxDFS
4: BeeGFS
5: ParaView
6: VTKm