Young Seo Lee

Young Seo Lee

Auto-generated profile, based on publicly available information

Young Seo Lee

Department of Computer Science and Engineering, Korea University, Seoul, South Korea

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
14 March 2022
Verified
Conference Paper
Closed access
YLYoung Seo Lee
Young Seo Lee
GKGunjae Koo
Gunjae Koo
YGYoung-Ho Gong
Young-Ho Gong
SCSung Woo Chung
Sung Woo Chung
As DRAM process technology scales down and DRAM density continues to grow, DRAM errors have become a primary concern in modern data centers. Typically, data centers have adopted memory systems with a single error correction double error detection (SECDED) code. However, the SECDED code is not sufficient to satisfy DRAM reliability demands as memory systems get more vulnerable. Though the servers in data centers employ strong ECC schemes, such ECC schemes lead to substantial performance and/or storage overhead. In this paper, we propose Stealth ECC, a cost-effective memory protection scheme providing stronger error correctability than the conventional SECDED code, with negligible performance overhead and without storage overhead. Depending on the data-width (either narrow-width or full-width), Stealth ECC adaptively selects ECC schemes. For narrow-width values, Stealth ECC provides multi-bit error correctability by storing more parity bits in MSB side, instead of zeros. Furthermore, with bitwise interleaved data placement between x4 DRAM chips, Stealth ECC is robust to a single DRAM chip error for narrow-width values. On the other hand, for full-width values, Stealth ECC adopts the SECDED code, which maintains DRAM reliability comparable to the conventional SECDED code. As a result, thanks to the reliability improvement of narrow-width values, Stealth ECC enhances overall DRAM reliability, while incurring negligible performance overhead as well as no storage overhead. Our simulation results show that Stealth ECC reduces the probability of system failure (caused by DRAM errors) by 47.9%, on average, with only 0.9% performance overhead compared to the conventional SECDED code.

8 January 2021 in IEEE Embedded Systems Letters
Verified
Journal Article
Closed access
YLYoung Seo Lee
Young Seo Lee
ECEui-Young Chung
Eui-Young Chung
YGYoung-Ho Gong
Young-Ho Gong
SCSung Woo Chung
Sung Woo Chung
Layerwise quantized neural networks (QNNs), which adopt different precisions for weights or activations in a layerwise manner, have emerged as a promising approach for embedded systems. The layerwise QNNs deploy only required number of data bits for the computation (e.g., convolution of weights and activations), which in turn reduces computation energy compared to the conventional QNNs. However, the layerwise QNNs still cause a large amount of energy in the conventional memory systems, since memory accesses are not optimized for the required precision of each layer. To address this problem, we propose Quant-PIM , an energy-efficient processing-in-memory (PIM) accelerator for layerwise QNNs. Quant-PIM selectively reads only required data bits within a data word depending on the precision, by deploying the modified I/O gating logics in a 3-D stacked memory. Thus, Quant-PIM significantly reduces energy consumption for memory accesses. In addition, Quant-PIM improves the performance of layerwise QNNs. When the required precision is half of the weight (or activation) size or less, Quant-PIM reads two data blocks in a single read operation by exploiting the saved memory bandwidth from the selective memory access, thus providing higher compute-throughput. Our simulation results show that Quant-PIM reduces system energy by 39.1%~50.4% compared to the PIM system with 16-bit quantized precision, without accuracy loss.

12 June 2020 in IEEE Embedded Systems Letters
Verified
Journal Article
Closed access
CDCong Thuan Do
Cong Thuan Do
JCJeong Hwan Choi
Jeong Hwan Choi
YLYoung Seo Lee
Young Seo Lee
CKCheol Hong Kim
Cheol Hong Kim
SCSung Woo Chung
Sung Woo Chung
Convolutional neural networks (CNNs) are one of the most popular machine learning algorithms. The convolutional layers, which account for the most execution time of CNNs, are implemented with matrix multiplication because the convolution operation performs dot products between filters and local regions of the input. On the other hand, GPUs with thousands of cores were proven to significantly accelerate matrix multiplication, compared to CPUs with a limited number of cores, especially for large matrices. However, the current memory architecture allows only one row access at a time so that multiple accesses are necessary to read the column data of the second matrix, thus slowing down matrix multiplication. In this study, we adopt the monolithic 3-D integration for the GPU scratchpad memory, called monolithic 3-D integration (M3D) scratchpad memory (SPM), to enhance matrix multiplication. The M3D SPM allows one access to read the column data of the second matrix, similar to the case of the first matrix. The simulation results show that our M3D SPM improves the system performance by 46.3% for the 32×32 matrix multiplication, over the conventional 2-D SPM where the column data of the second matrix are read sequentially.

30 August 2021 in IEEE Access
Verified
Journal Article
Open access
JLJi Heon Lee
Ji Heon Lee
YLYoung Seo Lee
Young Seo Lee
JCJeong Hwan Choi
Jeong Hwan Choi
HAHussam Amrouch
Hussam Amrouch
JKJoonho Kong
Joonho Kong
SCSung Woo Chung
Sung Woo Chung
Monolithic 3D (M3D) integration reduces the wire length, which eventually improves energy efficiency and performance compared to 2D integration. However, 3D integration inevitably causes higher on-chip temperature compared to 2D integration due to the increased power density as well as worse heat dissipation. The high on-chip temperature may offset the benefits of the M3D microprocessors due to the following reasons: 1) high on-chip temperature increases leakage power, which degrades energy efficiency. 2) the actual clock frequency is limited at run-time by frequent dynamic thermal management (DTM) invocations. In this paper, for the first time, we explore the thermal feasibility (whether it is possible to achieve high energy efficiency and performance without exceeding threshold temperature) of the M3D microprocessors depending on cooling solutions. For the thermal feasibility study, we construct an integrated framework to investigate the thermal behaviors and thermal feasibility of different types of microprocessors (M3D, 2D, and through-silicon-via based 3D (TSV-3D)) with different cooling solutions. Our thermal-aware evaluation results show that the best configuration of the M3D microprocessors reduces average energy consumption by 27.6% compared to the 2D microprocessor at an iso-frequency (4.0GHz). In addition, at the highest clock frequencies satisfying both design and thermal constraints, the best configuration of the M3D microprocessors improves average system performance by 25.1% and 26.0% compared to the 2D and TSV-3D microprocessors, respectively.