Shuhao Zhang

ORCID: 0000-0002-9927-6925
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Advanced Database Systems and Queries
  • Cloud Computing and Resource Management
  • Distributed systems and fault tolerance
  • Advanced Data Storage Technologies
  • Data Stream Mining Techniques
  • Data Management and Algorithms
  • Parallel Computing and Optimization Techniques
  • Algorithms and Data Compression
  • Anomaly Detection Techniques and Applications
  • Peer-to-Peer Network Technologies
  • Advanced Clustering Algorithms Research
  • Time Series Analysis and Forecasting
  • Caching and Content Delivery
  • Nitric Oxide and Endothelin Effects
  • Distributed and Parallel Computing Systems
  • Smart Parking Systems Research
  • Network Security and Intrusion Detection
  • Virtual Reality Applications and Impacts
  • IoT and Edge/Fog Computing
  • Software-Defined Networks and 5G
  • Underwater Vehicles and Communication Systems
  • Web Data Mining and Analysis
  • Spam and Phishing Detection
  • Vehicle License Plate Recognition
  • Radio Frequency Integrated Circuit Design

Huazhong University of Science and Technology
2025

Nanyang Technological University
2013-2024

Harbin Institute of Technology
2023-2024

Center for Life Sciences
2023-2024

Tsinghua University
2023-2024

Xi’an Jiaotong-Liverpool University
2023-2024

Jiangsu University
2024

Singapore University of Technology and Design
2021-2023

Beijing Advanced Sciences and Innovation Center
2023

Wuhan University
2021-2023

<h3>Background</h3> CD8<sup>+</sup> T cells differentiate into exhausted status within tumors, including hepatocellular carcinoma (HCC), which constitutes a solid barrier to effective anti-tumor immunity. A detailed characterization of and their prognostic value in HCC is lacking. <h3>Methods</h3> We collected fresh tumor tissues with adjacent non-tumor liver blood specimens 56 patients, as well archived samples from two independent cohorts patients (<i>n</i> = 358 <i>n</i> 254), who...

10.1186/s40425-019-0814-7 article EN cc-by Journal for ImmunoTherapy of Cancer 2019-11-29

This review aims to provide a systematic overview of high-energy-density anode-free lithium pouch batteries, focusing on the research progress in recent years and predicting subsequent development directions.

10.1039/d3ee01599h article EN Energy & Environmental Science 2023-01-01

Architecture designers tend to integrate both CPUs and GPUs on the same chip deliver energy-efficient designs. It is still an open problem effectively leverage advantages of integrated architectures. In this work, we port 42 programs in Rodinia, Parboil, Polybench benchmark suites analyze co-running behaviors these AMD Intel We find that performance not always better than running program only with or GPUs. Among programs, eight can benefit from co-running, while 24 using seven achieve best...

10.1109/tpds.2016.2586074 article EN IEEE Transactions on Parallel and Distributed Systems 2016-06-29

Recently, there have been some emerging processor designs that the CPU and GPU (Graphics Processing Unit) are integrated in a single chip share Last Level Cache (LLC). However, main memory bandwidth of such coupled CPU-GPU architectures can be much lower than discrete GPU. As result, current query co-processing paradigms severely suffer from stalls. In this paper, we propose novel in-cache paradigm for On-Line Analytical (OLAP) databases on architectures. Specifically, adapt CPU-assisted...

10.14778/2735496.2735497 article EN Proceedings of the VLDB Endowment 2014-12-01

Driven by the rapid hardware development of parallel CPU/GPU architectures, we have witnessed emerging relational query processing techniques and implementations on those architectures. However, most are not portable across different because they usually developed from scratch target at a specific architecture. This paper proposes kernel-adapter based design (OmniDB), yet efficient processor OmniDB attempts to develop an extensible kernel (qKernel) abstract model for leverage...

10.14778/2536274.2536319 article EN Proceedings of the VLDB Endowment 2013-08-01

MapReduce, originally developed by Google for search applications, has recently become a popular programming framework parallel and distributed environments. This paper presents an energy-efficient architecture design MapReduce on Field Programmable Gate Arrays (FPGAs). The major goal is to enable users program FPGAs with simple interfaces, meanwhile embrace automatic performance optimizations within the framework. Compared other processors like CPUs GPUs, are (re-)programmable hardware have...

10.1109/tpds.2016.2537805 article EN IEEE Transactions on Parallel and Distributed Systems 2016-03-03

The proliferation of 3D scanning technology, particularly within autonomous driving, has led to an exponential increase in the volume Point Cloud (PC) data. Given rich semantic information contained PC data, deep learning models are commonly employed for tasks such as object queries. However, current query systems that support data types do not process queries on information. Consequently, there is a notable gap research regarding efficiency invoking each query, especially when dealing with...

10.1145/3709702 article EN other-oa Proceedings of the ACM on Management of Data 2025-02-10

Driven by the rapidly increasing demand for handling real-time data streams, many stream processing (DSP) systems have been proposed. Regardless of different architectures those DSP systems, they are mostly aiming at scaling out using a cluster commodity machines and built around number key design aspects: a) pipelined with message passing, b) on-demand parallelism, c) JVM based implementation. However, there lacks study on aspects modern scale-up architectures, where more CPU cores being...

10.1109/icde.2017.119 article EN 2017-04-01

We introduce BriskStream, an in-memory data stream processing system (DSPSs) specifically designed for modern shared-memory multicore architectures. BriskStream's key contribution is execution plan optimization paradigm, namely RLAS, which takes relative-location (i.e., NUMA distance) of each pair producer-consumer operators into consideration. propose a branch and bound based approach with three heuristics to resolve the resulting nontrivial problem. The experimental evaluations demonstrate...

10.1145/3299869.3300067 preprint EN Proceedings of the 2022 International Conference on Management of Data 2019-06-18

Network function virtualization (NFV) is a promising technology that decouples network functions from hardware. Connecting virtual (VNFs) in series to form service chain (SFC) can flexibly orchestrate and expand functions. However, there are higher availability requirements for SFCs. This paper aims solve the SFC placement problem under resource constraints. proposes sideway cross (SC) backup model, which considers of both VNFs physical machines (PMs) data center. The SC model cross-arranges...

10.1109/tnsm.2022.3145103 article EN IEEE Transactions on Network and Service Management 2022-01-21

Stream processing prevails and SQL query on streams has become one of the most popular application scenarios. For example, in 2021, global number active IoT endpoints reaches 12.3 billion. Unfortunately, increasing scale data strict user requests place much pressure existing stream systems, requiring high throughput with low latency. To further improve performance current we propose a compression-based engine, called CompressStreamDB, which enables adaptive fine-grained directly compressed...

10.1109/icde55515.2023.00038 article EN 2022 IEEE 38th International Conference on Data Engineering (ICDE) 2023-04-01

In big cities, there are plenty of parking spaces, but we often find nowhere to park. For example, New York has 1.4 million cars and 4.4 on-street it is still not easy a place near our destination, especially during peak hours. The reason the lack prediction behavior. If could provide behavior in advance, can ease this problem that affects human well-being. We observe lots have periodic patterns, which an important factor for prediction. Unfortunately, existing work ignores such patterns...

10.24963/ijcai.2020/610 article EN 2020-07-01

Exploring the sharing opportunities among multiple stream queries is crucial for high-performance processing. Modern processing necessitates accelerating by utilizing heterogeneous coprocessors, such as GPUs, and this has shown to be an effective method. Emerging CPU-GPU integrated architectures 6integrate CPU GPU on same chip eliminate PCI-e bandwidth bottleneck. Such a novel architecture provides new improving multi-query performance in but not been fully explored existing systems. We...

10.1109/tpds.2021.3066407 article EN IEEE Transactions on Parallel and Distributed Systems 2021-03-17

SAP Event Stream Processor (ESP) platform aims at delivering real-time stream processing and analytics in many time-critical areas such as Capital Markets, Internet of Things (IoT) Data Center Intelligence. ESP allows users to realize complex event (CEP) the form pattern queries. In this paper, we present MOTTO - a multi-query optimizer order improve performance concurrent This is motivated by observations that real-world applications usually have queries working on same data streams,...

10.1109/icde.2017.166 article EN 2017-04-01

Data stream processing systems (DSPSs) enable users to express and run applications continuously process data streams. To achieve realtime analytics, recent researches keep focusing on optimizing the system latency throughput. Witnessing great achievements in computer architecture community, researchers practitioners have investigated potential of adoption hardware-conscious by better utilizing modern hardware capacity DSPSs. In this paper, we conduct a systematic survey work field,...

10.1145/3385658.3385662 article EN ACM SIGMOD Record 2020-02-25

There are plenty of parking spaces in big cities, but we often find nowhere to park. For example, New York has 1.4 million cars and 4.4 on-street spaces, it is still not easy a place near our destination, especially during peak hours. The reason the lack prediction behavior. If could provide behavior advance, can ease this problem that affects human well-being. We observe lots have periodic patterns, which an important factor for prediction. Unfortunately, existing work ignores such patterns...

10.1109/tkde.2021.3070202 article EN IEEE Transactions on Knowledge and Data Engineering 2021-03-31

In the burgeoning realm of Internet Things (IoT) applications on edge devices, data stream compression has become increasingly pertinent. The integration added overhead and limited hardware resources these devices calls for a nuanced software-hardware co-design. This paper introduces CStream, pioneering framework crafted parallelizing multicore devices. CStream grapples with distinct challenges delivering high ratio, throughput, low latency, energy consumption. Notably, distinguishes itself...

10.1109/tkde.2024.3386862 article EN IEEE Transactions on Knowledge and Data Engineering 2024-04-19

Double stranded DNA can adopt different forms, the so-called A-, B-, and Z-DNA, which play biological roles. In this work, thermodynamic kinetic parameters for base-pair closing opening in A-DNA B-DNA were calculated by all-atom molecular dynamics simulations at temperatures. The of base pair good agreement with experimental results. free energy barrier breaking a single stack results from enthalpy increase $\ensuremath{\Delta}H$ caused disruption hydrogen bonding base-stacking interactions,...

10.1103/physreve.103.042409 article EN Physical review. E 2021-04-14

The intra-window join (IaWJ), i.e., joining two input streams over a single window, is core operation in modern stream processing applications. This paper presents the first comprehensive study on parallelizing IaWJ multicore architectures. In particular, we classify algorithms into lazy and eager execution approaches. For each approach, there are further design aspects to consider, including different methods partitioning schemes, leading large space. Our results show that none of always...

10.1145/3448016.3452793 article EN Proceedings of the 2022 International Conference on Management of Data 2021-06-09

Transactional stream processing engines (TSPEs) differ significantly in their designs, but all rely on non- adaptive scheduling strategies for concurrent state transactions. Subsequently, none exploit multicore parallelism to its full potential due complex workload dependencies. This paper introduces MorphStream, which adopts a novel approach by decomposing into three dimensions and then strives make the right decision along each dimension, based analyzing trade-offs under varying...

10.1145/3588913 article EN Proceedings of the ACM on Management of Data 2023-05-26

Data Stream Clustering (DSC) plays an important role in mining continuous and unlabeled data streams real-world applications. Over the last decades, numerous DSC algorithms have been proposed with promising clustering accuracy efficiency. Despite significant differences among existing algorithms, they are commonly built around four key design aspects: summarizing structure, window model, outlier detection mechanism, offline refinement strategy. However, there is a lack of empirical studies...

10.1145/3589307 article EN Proceedings of the ACM on Management of Data 2023-06-13
Coming Soon ...