Zihan Liu

Zihan Liu

Auto-generated profile, based on publicly available information

Zihan Liu

Shanghai Jiao Tong University, Shanghai, China

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
22 February 2022
Verified
Conference Paper
Open access
ZLZihan Liu
Zihan Liu
JLJingwen Leng
Jingwen Leng
ZZZhihui Zhang
Zhihui Zhang
QCQuan Chen
Quan Chen
CLChao Li
Chao Li
MGMinyi Guo
Minyi Guo
Deep learning (DL) models have achieved great success in many application domains. As such, many industrial companies such as Google and Facebook have acknowledged the importance of multi-tenant DL services. Although the multi-tenant service has been studied in conventional workloads, it is not been deeply studied on deep learning service, especially on general-purpose hardware. In this work, we systematically analyze the opportunities and challenges of providing multi-tenant deep learning services on the general-purpose CPU architecture from the aspects of scheduling granularity and code generation. We propose an adaptive granularity scheduling scheme to both guarantee resource usage efficiency and reduce the scheduling conflict rate. We also propose an adaptive compilation strategy, by which we can dynamically and intelligently pick a program with proper exclusive and shared resource usage to reduce overall interference-induced performance loss. Compared to the existing works, our design can serve more requests under the same QoS target in various scenarios (e.g., +71%, +62%, +45% for light, medium, and heavy workloads, respectively), and reduce the averaged query latency by 50%.


27 April 2024
Verified
Conference Paper
Open access
ZLZihan Liu
Zihan Liu
WNWentao Ni
Wentao Ni
JLJingwen Leng
Jingwen Leng
YFYu Feng
Yu Feng
CGCong Guo
Cong Guo
YZYuhao Zhu
Yuhao Zhu
Approximate nearest neighbor (ANN) search is a widely applied technique in modern intelligent applications, such as recommendation systems and vector databases. Therefore, efficient and high-throughput execution of ANN search has become increasingly important. In this paper, we first characterize the state-of-the-art product quantization-based method of ANN search and identify a significant source of inefficiency in the form of unnecessary pairwise distance calculations and accumulations. To improve efficiency, we propose Juno, an end-to-end ANN search system that adopts a carefully designed sparsity- and locality-aware search algorithm. We also present an efficient hardware mapping that utilizes ray tracing cores in modern GPUs with pipelined execution on tensor cores to execute our sparsity-aware ANN search algorithm. Our evaluations on four datasets from 1 to 100 million search points demonstrate 2.2×-8.5× improvements in search throughput. Moreover, our algorithmic enhancements alone achieve a maximal 2.6× improvement on the hardware without the acceleration of the RT core.

20 November 2024 in ACM Transactions on Architecture and Code Optimization
Verified
Journal Article
Open access
YFYu Feng
Yu Feng
WLWeikai Lin
Weikai Lin
ZLZihan Liu
Zihan Liu
JLJingwen Leng
Jingwen Leng
MGMinyi Guo
Minyi Guo
YZYuhao Zhu
Yuhao Zhu
Neural Radiance Field (NeRF) has emerged as a promising alternative for photorealistic rendering. Despite recent algorithmic advancements, achieving real-time performance on today’s resource-constrained devices remains challenging. In this article, we identify the primary bottlenecks in current NeRF algorithms and introduce a unified algorithm-architecture co-design, Potamoi , designed to accommodate various NeRF algorithms. Specifically, we introduce a runtime system featuring a plug-and-play algorithm, SpaRW , which significantly reduces the per-frame computational workload and alleviates compute inefficiencies. Furthermore, our unified streaming pipeline coupled with customized hardware support effectively tames both SRAM and DRAM inefficiencies by minimizing repetitive DRAM access and completely eliminating SRAM bank conflicts. When evaluated against a baseline utilizing a dedicated DNN accelerator, our framework demonstrates a speedup and energy reduction of 53.1× and 67.7×, respectively, all while maintaining high visual quality with less than a 1.0 dB reduction in peak signal-to-noise ratio.

20 June 2025
Verified
Conference Paper
Open access
YFYu Feng
Yu Feng
WLWeikai Lin
Weikai Lin
YCYuge Cheng
Yuge Cheng
ZLZihan Liu
Zihan Liu
JLJingwen Leng
Jingwen Leng
YZYuhao Zhu
Yuhao Zhu

1 March 2021
Verified
Conference Paper
Closed access
ZLZihan Liu
Zihan Liu
QLQizhong Li
Qizhong Li
SMShengming Mei
Shengming Mei
MHMiaohua Huang
Miaohua Huang
The roadside LiDAR(light detection and ranging) sensors can obtain relevant information about all roadway users by collecting real-time 3D point cloud data of surrounding objects. In order to analyze the point cloud data of roadway users, background filtering and object detection are essential steps. This paper proposes a background filtering method based on point correlation by KDTree neighborhood searching and an adaptive threshold Euclidean clustering method. In order to improve the detection speed, background filtering will be performed on the original point cloud data before Euclidean clustering, so as to delete a large number of useless background points, and the clustering results will be distinguished by cuboid frame markers. The experimental results show that the single-frame point cloud clustering speed has increased by 31.3%, the clustering speed has been significantly improved.