Jaeyoung Do

Jaeyoung Do

Auto-generated profile, based on publicly available information

Jaeyoung Do

Seoul National University, Seoul, Republic of Korea

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 March 2022 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
JKJi-Hoon Kim
Ji-Hoon Kim
YPYeo-Reum Park
Yeo-Reum Park
JDJaeyoung Do
Jaeyoung Do
SJSoo-Young Ji
Soo-Young Ji
JKJoo-Young Kim
Joo-Young Kim
$K$ -nearest neighbor search is one of the fundamental tasks in various applications and the hierarchical navigable small world (HNSW) has recently drawn attention in large-scale cloud services, as it easily scales up the database while offering fast search. On the other hand, a computational storage device (CSD) that combines programmable logic and storage modules on a single board becomes popular to address the data bandwidth bottleneck of modern computing systems. In this paper, we propose a computational storage platform that can accelerate a large-scale graph-based nearest neighbor search algorithm based on SmartSSD CSD. To this end, we modify the algorithm more amenable on the hardware and implement two types of accelerators using HLS- and RTL-based methodology with various optimization methods. In addition, we scale up the proposed platform to have 4 SmartSSDs and apply graph parallelism to boost the system performance further. As a result, the proposed computational storage platform achieves 75.59 query per second throughput for the SIFT1B dataset at 258.66W power dissipation, which is 12.83x and 17.91x faster and 10.43x and 24.33x more energy efficient than the conventional CPU-based and GPU-based server platform, respectively. With multi-terabyte storage and custom acceleration capability, we believe that the proposed computational storage platform is a promising solution for cost-sensitive cloud datacenters.

11 May 2024
Verified
Conference Paper
Open access
JHJiyeon Han
Jiyeon Han
JPJimin Park
Jimin Park
JHJinyoung Huh
Jinyoung Huh
UOUran Oh
Uran Oh
JDJaeyoung Do
Jaeyoung Do
DKDaehee Kim
Daehee Kim
While clinical notes are essential to the field of healthcare, they pose several challenges for clinicians since it is difficult to write down medical information, review prior notes, and extract the desired information at the same time while examining a patient. Thus, we designed a system that can automatically generate clinical notes from dialogues between patients and clinicians and provide specific information upon clinicians’ query using a Large Language Model (LLM) both in real-time. To explore how this system can be used to support clinicians in practice, we conducted an interview with six clinicians followed by a design probe study with the current version of our system for feedback. Findings suggest that our system has the potential to enable clinicians to write and access clinical notes and examine the patients simultaneously with reduced cognitive loads and increased efficiency and accuracy.

18 February 2021 in The VLDB Journal
Verified
Journal Article
Closed access
PDF file
JDJaeyoung Do
Jaeyoung Do
IPIvan Luiz Picoli
Ivan Luiz Picoli
DLDavid Lomet
David Lomet
PBPhilippe Bonnet
Philippe Bonnet

11 January 2023 in ACM Transactions on Storage
Verified
Journal Article
Open access
HLHuaicheng Li
Huaicheng Li
MPMartin L. Putra
Martin L. Putra
RSRonald Shi
Ronald Shi
FKFadhil I. Kurnia
Fadhil I. Kurnia
XLXing Lin
Xing Lin
HGHaryadi S. Gunawi
Haryadi S. Gunawi
Predictable latency on flash storage is a long-pursuit goal, yet unpredictability stays due to the unavoidable disturbance from many well-known SSD internal activities. To combat this issue, the recent NVMe IO Determinism (IOD) interface advocates host-level controls to SSD internal management tasks. Although promising, challenges remain on how to exploit it for truly predictable performance. We present IODA , 1 an I/O deterministic flash array design built on top of small but powerful extensions to the IOD interface for easy deployment. IODA exploits data redundancy in the context of IOD for a strong latency predictability contract. In IODA , SSDs are expected to quickly fail an I/O on purpose to allow predictable I/Os through proactive data reconstruction. In the case of concurrent internal operations, IODA introduces busy remaining time exposure and predictable-latency-window formulation to guarantee predictable data reconstructions. Overall, IODA only adds five new fields to the NVMe interface and a small modification in the flash firmware while keeping most of the complexity in the host OS. Our evaluation shows that IODA improves the 95–99.99 th latencies by up to 75×. IODA is also the nearest to the ideal, no disturbance case compared to seven state-of-the-art preemption, suspension, GC coordination, partitioning, tiny-tail flash controller, prediction, and proactive approaches.

24 September 2026 in Proceedings of the ACM on Management of Data
Verified
Journal Article
Open access
GCGeon Choi
Geon Choi
Hoeun LeeHoeun Lee
Hoeun Lee
JDJaeyoung Do
Jaeyoung Do
Vector databases (VectorDBs) are a core component of modern retrieval systems, including Retrieval-Augmented Generation (RAG), where efficient Approximate Nearest Neighbor Search (ANNS) is critical. Among ANNS algorithms, Hierarchical Navigable Small World (HNSW) graphs are widely adopted for their strong recall-latency trade-off. However, configuring HNSW remains challenging: its hyperparameters jointly affect search quality, latency, build time, and index size in nonlinear ways, while production deployments impose strict resource and tuning-time constraints. We study HNSW hyperparameter tuning from a systems perspective and show that its configuration space exhibits strong structural regularities. Specifically, we identify monotonic, dominant unimodal, and separable relationships among search-time and construction-time parameters, which induce feasibility boundaries under performance and resource constraints. Building on this insight, we propose CHAT, a constraint-aware tuning framework for HNSW. Unlike generic black-box optimizers, CHAT exploits HNSW-specific structure to perform deterministic, sample-efficient search and prune resource-infeasible configurations before full index construction. Across multiple datasets and HNSW-based vector search engines, CHAT identifies configurations that maximize recall or throughput while satisfying constraints on accuracy, latency, build time, index size, and tuning budget. Compared to strong baselines, CHAT achieves up to 45% higher throughput or 11% higher recall, and converges up to 44× faster. These results show that principled, structure-aware tuning enables efficient and robust HNSW deployment beyond generic black-box optimization.

1 January 2023
Verified
Conference Paper
Open access
PDF file
SASarthak Ahuja
Sarthak Ahuja
MKMohammad Kachuee
Mohammad Kachuee
FSFatemeh Sheikholeslami
Fatemeh Sheikholeslami
WLWeiqing Liu
Weiqing Liu
JDJaeyoung Do
Jaeyoung Do
Sarthak Ahuja, Mohammad Kachuee, Fatemeh Sheikholeslami, Weiqing Liu, Jaeyoung Do. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track). 2023.