检索结果-南通市图书馆

Learned Distributed Query Optimizer:Architecture and Challenges

在线全文

学校读者我要写书评

暂无评论

ZTE Communications 2024年第2期22卷 49-54页

作者： GAO Jun han yinjun LIN Yang MIAO Hao XU Mo Peking University Beijing 100871China ZTE Corporation Shenzhen 518057China

The query processing in distributed database management systems(DBMS)faces more challenges,such as more operators,and more factors in cost models and meta-data,than that in a single-node DMBS,in which query optimization is already an NP-hard *** query optimizers(mainly in the single-node DBMS)receive attention due to its capability to capture data distributions and flexible ways to avoid hard-craft rules in refinement and adaptation to new *** this paper,we focus on extensions of learned query optimizers to distributed ***,we propose one possible but general architecture of the learned query optimizer in the distributed context and highlight differences from the learned optimizer in the single-node *** addition,we discuss the challenges and possible solutions.

关键词： distributed query processing query optimization learned query optimizer

DDUC:an erasure-coded system with decoupled data updating and coding

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2023年第5期24卷 716-730页

作者： Yaofeng TU Rong XIAO yinjun han Zhenghua CHEN Hao JIN Xuecheng QI Xinyuan SUN State Key Laboratory of Mobile Network and Mobile Multimedia Technology Shenzhen 518000China ZTE Corporation Nanjing 210000China

In distributed storage systems,replication and erasure code(EC)are common methods for data *** with replication,EC has better storage efficiency,but suffers higher overhead in ***,consistency and reliability problems caused by concurrent updates bring new challenges to applications of *** works focus on optimizing the EC solution,including algorithm optimization,novel data update method,and so on,but lack the solutions for consistency and reliability *** this paper,we introduce a storage system that decouples data updating and EC encoding,namely,decoupled data updating and coding(DDUC),and propose a data placement policy that combines replication and parity *** the(N,M)EC system,the data are placed as N groups of M+1 replicas,and redundant data blocks of the same stripe are placed in the parity nodes,so that the parity nodes can autonomously perform local EC *** on the above policy,a two-phase data update method is implemented in which data are updated in replica mode in phase 1,and the EC encoding is done independently by parity nodes in phase *** solves the problem of data reliability degradation caused by concurrent updates while ensuring high concurrency *** also uses persistent memory(PMem)hardware features of the byte addressing and eight-byte atomic write to implement a lightweight logging mechanism that improves performance while ensuring data *** results show that the concurrent access performance of the proposed storage system is 1.70–3.73 times that of the state-of-the-art storage system Ceph,and the latency is only 3.4%–5.9%that of Ceph.

关键词： Concurrent update High reliability Erasure code Consistency Distributed storage system

在线全文

学校读者我要写书评

暂无评论

UStore:面向新型硬件的统一存储系统

计算机研究与发展 2023年第3期60卷 525-538页

作者：屠要峰韩银俊金浩陈正华陈兵南京航空航天大学计算机科学与技术学院南京211106 中兴通讯股份有限公司南京210012

数据规模的爆发式增长使得分布式存储广泛应用,长期以来分布式存储直接使用本地文件系统访问本地存储资源,随着高性能NVMe SSD、持久内存(persistent memory,PMEM)、异构加速设备的出现,本地文件系统难以发挥新型器件的特性和性能优势.... 详细信息

数据规模的爆发式增长使得分布式存储广泛应用,长期以来分布式存储直接使用本地文件系统访问本地存储资源,随着高性能NVMe SSD、持久内存(persistent memory,PMEM)、异构加速设备的出现,本地文件系统难以发挥新型器件的特性和性能优势.现有很多研究针对SSD或者PMEM的硬件特性在软件层面进行优化和改进.然而,这些工作兼容性和扩展性差,不能灵活适配硬件环境变化,缺乏面向新型硬件的统一解决方案.提出了一种兼容多种存储介质的统一存储系统UStore,可根据业务场景灵活选择存储介质,并针对PMEM、KVS加速卡、NVMe SSD等典型硬件进行组合设计优化,充分利用其硬件特性,满足多层次的业务需求;通过一种与物理存储介质形态解耦的元数据设计,使其适应不同硬件的性能和原子更新能力差异,实现灵活的元数据存储策略;通过高效的数据管理机制和更新策略,实现无日志的数据原子写保证,消除了现有系统的写放大以及性能抖动等问题.实验结果表明,相比于BlueStore,UStore的4KB随机读性能提升了3.2倍,4KB随机写提高了8.2倍,3种典型硬件组合下UStore表现出与之相匹配的数据访问特征,充分发挥了存储硬件的特性和性能.

关键词：持久性内存 NVMe SSD 分布式存储元数据原子写

一种持久性内存文件系统数据页的混合管理机制

在线全文

学校读者我要写书评

暂无评论

计算机研究与发展 2020年第2期57卷 281-290页

作者：陈游旻朱博弘韩银俊屠要峰舒继武清华大学计算机科学与技术系北京100084 中兴通讯股份有限公司南京210012

英特尔于2019年4月正式发布基于3D-Xpoint技术的傲腾持久性内存(Optane DC persistent memory),这为构建高效的持久性内存存储系统提供了新的机遇.然而,现有的存储系统软件并不能很好地利用其字节寻址特性,持久性内存性能很难充分发挥.... 详细信息

英特尔于2019年4月正式发布基于3D-Xpoint技术的傲腾持久性内存(Optane DC persistent memory),这为构建高效的持久性内存存储系统提供了新的机遇.然而,现有的存储系统软件并不能很好地利用其字节寻址特性,持久性内存性能很难充分发挥.提出一种文件系统数据页的混合管理机制HDPM,通过选择性使用写时复制机制和日志结构管理文件数据,充分发挥持久性内存字节可寻址特性,从而避免了传统单一模式在非对齐写或者小写造成的写放大问题.为避免影响读性能,HDPM引入逆向扫描机制,实现日志结构重构数据页时不引入额外数据拷贝.HDPM还提出一种多重垃圾回收机制进行日志清理.当单个日志结构过大时,通过读写流程主动回收日志结构;当持久性内存空间受限时,则通过后台线程使用免锁机制异步释放日志空间.实验显示,HDPM相比于NOVA文件系统,单线程写延迟降低达58%,且读延迟不受影响;Filebench多线程测试显示,HDPM相比于NOVA提升吞吐率33%.

关键词：持久性内存文件系统写时复制日志结构垃圾回收

基于持久性内存和SSD的后端存储MixStore

在线全文

学校读者我要写书评

暂无评论

计算机研究与发展 2021年第2期58卷 406-417页

作者：屠要峰陈正华韩银俊陈兵关东海南京航空航天大学计算机科学与技术学院南京211106 中兴通讯股份有限公司南京210012

持久性内存(persistent memory,PMEM)同时具备内存的低时延字节寻址和磁盘的持久化特性,将对现有软件架构体系产生革命性的变化和深远的影响.分布式存储在云计算和数据中心得到了广泛的应用,然而现有的以Ceph BlueStore为代表的后端存... 详细信息

持久性内存(persistent memory,PMEM)同时具备内存的低时延字节寻址和磁盘的持久化特性,将对现有软件架构体系产生革命性的变化和深远的影响.分布式存储在云计算和数据中心得到了广泛的应用,然而现有的以Ceph BlueStore为代表的后端存储引擎是面向传统机械盘和固态硬盘(solid state disk,SSD)设计的,其原有的优化设计机制不适合PMEM特性优势的发挥.提出了一种基于持久性内存和SSD的后端存储MixStore,通过易失区段标记和待删除列表技术实现了适用于持久性内存的并发跳表,用于替代RocksDB实现元数据管理机制,在保证事务一致性的同时,消除了BlueStore的compaction所引发的性能抖动等问题,同时提升元数据的并发访问性能;通过结合元数据管理机制的数据对象存储优化设计,把非对齐的小数据对象存放在PMEM中,把对齐的大块数据对象存储在SSD上,充分发挥了PMEM的字节寻址、持久性特性和SSD的大容量低成本优势,并结合延迟写入和CoW(copy-on-write)技术实现数据更新策略优化,消除了BlueStore的WAL日志引起的写放大,提升小数据写入性能.测试结果表明,在同样的硬件环境下,相比BlueStore,MixStore的写吞吐提升59%,写时延降低了37%,有效地提升了系统的性能.

关键词：持久性内存并发跳表后端存储混合存储 BlueStore存储引擎

在线全文

学校读者我要写书评

暂无评论

AI辅助的智能分级存储系统

信息化研究 2022年第6期48卷 52-58页

作者：高洪韩银俊南京中兴软件有限责任公司南京210012 南京中兴新软件有限责任公司南京210012

分布式存储系统中常用的冗余模式有副本和纠删码(EC)两种。副本模式存储效率低,EC模式读写性能不高。文章研究了它们的优劣,结合普通硬件和新型NVMe SSD硬件,实施了一种智能的分级存储的分布式存储系统ATSS,热数据以副本模式存储在高性... 详细信息

分布式存储系统中常用的冗余模式有副本和纠删码(EC)两种。副本模式存储效率低,EC模式读写性能不高。文章研究了它们的优劣,结合普通硬件和新型NVMe SSD硬件,实施了一种智能的分级存储的分布式存储系统ATSS,热数据以副本模式存储在高性能硬件上,以保障高并发高吞吐的读写,冷数据转换为EC模式存储在普通硬件上,节约了系统磁盘空间的同时保障了数据的可靠性。同时通过AI辅助判断需要进行EC编码的数据块,避免了热数据被编码迁移到普通硬件,使得系统在性能和空间效率之间取得了平衡。实验显示,相较于主流的分布式存储Ceph,本系统读性能提升2倍以上,而延迟仅为Ceph的4%左右。

关键词：人工智能纠删码分级分布式存储系统

在线全文

学校读者我要写书评

暂无评论

NVMe优化的用户态文件系统

信息化研究 2022年第5期48卷 44-49,60页

作者：高洪韩银俊南京中兴软件有限责任公司南京210012 南京中兴新软件有限责任公司南京210012

NVMe接口协议的引入使得固态硬盘(SSD)存储设备的性能得到显著提升。但是,传统的内核态IO路径阻碍了NVMe SSD设备的性能发挥。文中提出了一种面向NVMe SSD的用户态文件系统UMFS,通过用户态多进程共享内存的设计,实现了多个应用程序共享... 详细信息

NVMe接口协议的引入使得固态硬盘(SSD)存储设备的性能得到显著提升。但是,传统的内核态IO路径阻碍了NVMe SSD设备的性能发挥。文中提出了一种面向NVMe SSD的用户态文件系统UMFS,通过用户态多进程共享内存的设计,实现了多个应用程序共享访问,减少对SSD的访问量;使用面向NVMe的无日志数据布局,多粒度IO队列分离和NUMA感知的访问技术,提升系统的性能和吞吐量。实验显示,相较于传统文件系统,在CDN场景下UMFS性能提升23%以上,在小文件场景和读取密集场景下UMFS性能提升45%以上。

关键词：文件系统用户态 NVMe固态硬盘无日志

Recent advances in exploring new blood-based biomarkers for the early diagnosis of gastric cancer

在线全文

学校读者我要写书评

暂无评论

Science China Chemistry 2025年第1期68卷 35-45页

作者： Xinyu Peng Qian Ma Da han School of Life Science Shanghai University Institute of Molecular Medicine Shanghai Key Laboratory for Nucleic Acid Chemistry and Nanomedicine State Key Laboratory of Oncogenes and Related Genes Renji Hospital School of Medicine Shanghai Jiao Tong University Department of Research and Development Intellinosis Biotechnologies Co.Ltd.

Gastric cancer(GC) is a prevalent and lethal malignancy worldwide, and the 5-year survival rate is less than 30%. The absence of distinctive symptoms often leads to late-stage diagnosis, contributing to a grim prognosis. Early detection is pivotal for improving the survival rates and outcomes of GC patients. Traditional diagnostic methods, including gastroscopy and histopathological examination, are restricted in the early diagnosis of GC. Conventional protein biomarkers, such as carcinoembryonic antigen(CEA), carbohydrate antigen 19-9(CA19-9), and carbohydrate antigen 72-4(CA72-4), exhibit insufficient sensitivity and specificity for GC screening, which diminishes their clinical value. Recently, novel circulating biomarkers, such as circulating tumor cells(CTCs), circulating tumor cell DNA(ct DNA), and circulating free RNA(cf RNA), have emerged as promising candidates in the burgeoning field of liquid biopsy due to their superior sensitivity and specificity. This review focuses on the latest research on blood-based biomarkers for GC diagnosis and examines the clinical potentials and challenges associated with these emerging biomarkers.

关键词： gastric cancer early diagnosis blood

Secure beamforming and deployment design for rate-splitting multiple access-based UAV communications

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2025年第1期68卷 342-356页

作者： Bing WANG Xiaofeng TAO Shujun han Huici WU Kai YANG Zhu han National Engineering Research Center of Mobile Network Technologies Beijing University of Posts and Telecommunications Peng Cheng Laboratory Department of Electrical and Computer Engineering University of Houston

Rate-splitting multiple access(RSMA) has recently gained attention as a potential robust multiple access(MA)scheme for upcoming wireless networks. Given its ability to efficiently utilize wireless resources and design interference management strategies, it can be applied to unmanned aerial vehicle(UAV) networks to provide convenient services for large-scale access ground users. However, due to the line-of-sight(LoS) broadcast nature of UAV transmission, information is susceptible to eavesdropping in RSMA-based UAV networks. Moreover, the superposition of signals at the receiver in such networks becomes complicated. To cope with the challenge, we propose a two-user multi-input single-output(MISO) RSMA-based UAV secure transmission framework in downlink communication networks. In a passive eavesdropping scenario, our goal is to maximize the sum secrecy rate by optimizing the transmit beamforming and deployment location of the UAV-base station(UAV-BS),while considering quality-of-service(QoS) constraints, maximum transmit power, and flight space limitations. To address the non-convexity of the proposed problem, the optimization problem is first decoupled into two subproblems. Then, the successive convex approximation(SCA) method is employed to solve each subproblem using different propositions. In addition, an alternating optimization(AO)-based location RSMA(L-RSMA) beamforming algorithm is developed to implement joint optimization to obtain the suboptimal solution. Numerical results demonstrate that(1) the proposed L-RSMA scheme yields a28.97% higher sum secrecy rate than the baseline L-space division multiple access(SDMA) scheme;(2) the proposed L-RSMA scheme improves the security performance by 42.61% compared to the L-non-orthogonal multiple access(NOMA) scheme.

关键词： rate-splitting multiple access UAV communication precoding optimization deployment location physical layer security

同方期刊数据库