Xiaoyao Liang

Xiaoyao Liang

Auto-generated profile, based on publicly available information

Xiaoyao Liang

Shanghai Jiao Tong University

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
23 September 2023 in Nature Communications
Verified
Journal Article
Open access
PDF file
MXMaosong Xie
Maosong Xie
YJYueyang Jia
Yueyang Jia
CNChen Nie
Chen Nie
ZLZuheng Liu
Zuheng Liu
ATAlvin Tang
Alvin Tang
RYRui Yang
Rui Yang
Emerging data-intensive computation has driven the advanced packaging and vertical stacking of integrated circuits, for minimized latency and energy consumption. Yet a monolithic three-dimensional (3D) integrated structure with interleaved logic and high-density memory layers has been difficult to achieve due to challenges in managing the thermal budget. Here we experimentally demonstrate a monolithic 3D integration of atomically-thin molybdenum disulfide (MoS2) transistors and 3D vertical resistive random-access memories (VRRAMs), with the MoS2transistors stacked between the bottom-plane and top-plane VRRAMs. The whole fabrication process is integration-friendly (below 300 °C), and the measurement results confirm that the top-plane fabrication does not affect the bottom-plane devices. The MoS2transistor can drive each layer of VRRAM into four resistance states. Circuit-level modeling of the monolithic 3D structure demonstrates smaller area, faster data transfer, and lower energy consumption than a planar memory. Such platform holds a high potential for energy-efficient 3D on-chip memory systems.

1 February 2021 in Laboratory Investigation
Verified
Journal Article
Closed access
PDF file
JKJing Ke
Jing Ke
YSYiqing Shen
Yiqing Shen
YLYizhou Lu
Yizhou Lu
JDJunwei Deng
Junwei Deng
JWJason D. Wright
Jason D. Wright
FJFusong Jiang
Fusong Jiang

21 September 2021
Verified
Book Chapter
Closed access
PDF file
JKJing Ke
Jing Ke
YSYiqing Shen
Yiqing Shen
XLXiaoyao Liang
Xiaoyao Liang
DSDinggang Shen
Dinggang Shen

13 June 2023 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
CNChen Nie
Chen Nie
CTChenyu Tang
Chenyu Tang
JLJie Lin
Jie Lin
HHHuan Hu
Huan Hu
CLChenyang Lv
Chenyang Lv
Zhezhi HeZhezhi He
Zhezhi He
Processing-in-Memory (PIM) has been widely explored for accelerating data-intensive machine learning computation that mainly consists of general-matrix-multiplication (GEMM), by mitigating the burden of data movements and exploiting the ultra-high memory parallelism. The two mainstreams of PIM, the analog- and digital-type, have both been exploited in accelerating machine learning workloads by numerous outstanding prior works. Currently, the digital-PIM is increasingly favored due to the broader computing support and the avoidance of errors caused by intrinsic non-idealities, e.g., process variation. Nevertheless, it still lacks further optimization considering the characteristics of the GEMM computation, including better efficient data layout and scheduling, and the ability to handle the sparsity of activations at the bit-level. To boost the performance and efficiency of digital SRAM PIM, we propose the architecture called VSPIM that performs the computation in a bit-serial fashion, with unique support of vector-scalar computing pattern. The novelties of the VSPIM can be concluded as follows: 1) support bit-serial based scalar-vector computing via ingenious parallel bit-broadcasting; 2) refine the GEMM mapping strategy and computing pattern to enhance performance and efficiency; 3) powered by the introduced scalar-vector operation, the bit-sparsity of activation is leveraged to halt unnecessary computation to maximize efficiency and throughput. Our comprehensive evaluation shows that, compared to the state-of-the-art SRAM-based digital-PIM design (Neural Cache), VSPIM can significantly boost the performance and energy efficiency by up to $8.87\times$ and $4.81\times$ respectively, with negligible area overhead, upon multiple representative neural networks.

27 April 2024
Verified
Conference Paper
Open access
ZSZhuoran Song
Zhuoran Song
CQChunyu Qi
Chunyu Qi
FLFangxin Liu
Fangxin Liu
NJNaifeng Jing
Naifeng Jing
XLXiaoyao Liang
Xiaoyao Liang
Video Transformers (VidTs) have reached the forefront of accuracy in various video understanding tasks. Despite their remarkable achievements, the processing requirements for a large number of video frames still present a significant performance bottleneck, impeding their deployment to resource-constrained platforms. While accelerators meticulously designed for Vision Transformers (ViTs) have emerged, they may not be the optimal solution for VidTs, primarily due to two reasons. These accelerators tend to overlook the inherent temporal redundancy that characterizes VidTs, limiting their chance for further performance enhancement. Moreover, incorporating a sparse attention prediction module within these accelerators incurs a considerable overhead. To this end, we move our attention to the video CODEC, which is essential for video preprocessing and can be utilized to detect the temporal and spatial similarity in raw video frames, showcasing the potential of exploring the temporal and spatial redundancy in VidTs while avoiding significant costs on prediction. This paper proposes CMC, the first CODEC assisted algorithm-accelerator co-design framework (CMC) for VidT acceleration. Specifically, from the algorithm aspects, we offer CODEC-friendly inter- and intra-matrix prediction algorithms to identify the informative data on-the-fly. We then design a recovery algorithm so that we can safely skip the computation on non-informative data in the temporal and spatial domains and recover their results by copying the informative data's features to reserve accuracy. From the hardware aspects, we propose to augment the video CODEC to make it efficiently implement inter- and intra-matrix prediction algorithms with negligible costs. Additionally, we propose a specialized CMC architecture that includes a recovery engine with fine-grained buffer management to translate the computational saving in the algorithm to real speedup. Experiments show that CMC can achieve 2.1×, 8.8×, and 42.4× speedup over state-of-the-art ViT accelerator HeatViT, A100 GPU, and Xeon CPU with negligible accuracy loss.

10 July 2023 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
FLFangxin Liu
Fangxin Liu
WZWenbo Zhao
Wenbo Zhao
ZWZongwu Wang
Zongwu Wang
YCYongbiao Chen
Yongbiao Chen
XLXiaoyao Liang
Xiaoyao Liang
LJLi Jiang
Li Jiang
Resistive Random-Access-Memory (ReRAM) crossbar is one of the most promising neural network accelerators, thanks to its in-memory and in-situ analog computing abilities for Matrix Multiplication-and-Accumulations (MACs). The key limitations are: 1) the number of rows and columns of ReRAM cells for concurrent execution of MACs is constrained, resulting in limited in-memory computing throughput; 2) the cost of high-precision analog-to-digital (A/D) conversions that can offset the efficiency and performance benefits of ReRAM-based Process-In-Memory (PIM). Meanwhile, it is challenging to deploy Deep Neural Network (DNN) models with a large model size in the crossbar since the sparsity of DNNs cannot be effectively exploited in the crossbar structure, especially the sparsity in the activation. As a countermeasure, we develop a novel ReRAM-based PIM accelerator, namely ERA-BS, which pays attention to the correlation between the bit-level sparsity (in both weights and activations) and the performance of the ReRAM-based crossbar. We propose a superior bit-flip scheme combined with the exponent-based quantization, which can adaptively flip the bits of the mapped DNNs to release redundant space without sacrificing the accuracy much or incurring much hardware overhead. Meanwhile, we design an architecture that can integrate the techniques to shrink the crossbar footprint to be used massively. We further propose a dynamic activation sparsity exploitation scheme in conjunction with the tightly coupled structure nature of the crossbar, including crossbar-aware activation pruning and ancillary run-time hardware support. In such a way, we exploit fine-grained sparsity weights (static) and activations (dynamic), respectively, to improve performance while reducing the energy consumption of computation with negligible overheads. Our experiments on a wide variety of networks show that compared to the well-known ReRAM-based PIM accelerator like “ISAAC”, ERA-BS can achieve up to $43\times$ , $78\times$ , and $73\times$ in terms of energy efficiency, area-efficiency, and throughput, respectively. Compared to the state-of-the-art ReRAM-based design “PIM-Prune”, ERA-BS can also achieve $5.3\times$ energy efficiency, $7.2\times$ area efficiency, and $32\times$ performance gain with a similar or even higher accuracy.