2025数据库论文阅读笔记

2026-01-26 14:48:09 +08:00
parent 870ab27f21
commit 77728a4c93
3 changed files with 152 additions and 1 deletions
--- a/references.bib
+++ b/references.bib
@@ -49,6 +49,11 @@
  archiveprefix = {arXiv}
 }

+@article{chancskqs,
+  title = {CSKQS: A Query System for Collective Spatial Keyword Queries},
+  author = {Chan, Harry Kai-Ho}
+}
+
@misc{chen2024llm,
  title = {An LLM Agent for Automatic Geospatial Data Analysis},
  author = {Chen, Yuxing and Wang, Weijie and Lobry, Sylvain and Kurtz, Camille},
@@ -75,6 +80,17 @@
  urldate = {2026-01-18}
 }

+@inproceedings{guo2025storage,
+  title = {A Storage Model with Fine-Grained In-Storage Query Processing for Spatio-Temporal Data},
+  booktitle = {2025 IEEE 41st International Conference on Data Engineering (ICDE)},
+  author = {Guo, Yang and Wang, Tianyu and Chen, Zizhan and Shao, Zili},
+  year = 2025,
+  pages = {669--682},
+  issn = {2375-026X},
+  doi = {10.1109/ICDE65448.2025.00056},
+  urldate = {2026-01-20}
+}
+
@article{hong2025hybrid,
  title = {A Hybrid Approach to Integrating Deterministic and Non-Deterministic Concurrency Control in Database Systems},
  author = {Hong, Yinhao and Zhao, Hongyao and Lu, Wei and Du, Xiaoyong and Chen, Yuxing and Pan, Anqun and Zheng, Lixiong},
@@ -127,6 +143,30 @@
  archiveprefix = {arXiv}
 }

+@inproceedings{liu2025hivq,
+  title = {HiVQ: A Real-time Interactive Visual Query System on Geospatial Big Data},
+  booktitle = {2025 IEEE 41st International Conference on Data Engineering (ICDE)},
+  author = {Liu, Zebang and Yang, Anran and Ma, Mengyu and Chen, Luo and Zhou, Jiali and Jing, Ning},
+  year = 2025,
+  pages = {4592--4595},
+  issn = {2375-026X},
+  doi = {10.1109/ICDE65448.2025.00360},
+  urldate = {2026-01-20}
+}
+
+@article{liu2025nalspatial,
+  title = {NALSpatial: A Natural Language Interface for Spatial Databases},
+  author = {Liu, Mengyi and Wang, Xieyang and Xu, Jianqiu and Lu, Hua and Tong, Yongxin},
+  year = 2025,
+  journal = {IEEE Transactions on Knowledge and Data Engineering},
+  volume = {37},
+  number = {4},
+  pages = {2056--2070},
+  issn = {1558-2191},
+  doi = {10.1109/TKDE.2025.3525587},
+  urldate = {2026-01-20}
+}
+
@phdthesis{lukai2024jiyujiqixuexidefenbushicunchuxitongxingnengyouhuayanjiu,
  type = {博士学位论文},
  title = {基于机器学习的分布式存储系统性能优化研究},
@@ -175,6 +215,46 @@ filename: 1024392702.nh}
  urldate = {2025-06-10}
 }

+@article{teng2025efficient,
+  title = {Efficient and Accurate Spatial Queries Using Lossy Compressed 3D Geometry Data},
+  author = {Teng, Dejun and Li, Zhaochuan and Peng, Zhaohui and Ma, Shuai and Wang, Fusheng},
+  year = 2025,
+  journal = {IEEE Transactions on Knowledge and Data Engineering},
+  volume = {37},
+  number = {5},
+  pages = {2472--2487},
+  issn = {1041-4347, 1558-2191, 2326-3865},
+  doi = {10.1109/TKDE.2025.3539729},
+  urldate = {2026-01-23},
+  copyright = {https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html}
+}
+
+@article{tong2022hufu,
+  title = {Hu-Fu: efficient and secure spatial queries over data federation},
+  author = {Tong, Yongxin and Pan, Xuchen and Zeng, Yuxiang and Shi, Yexuan and Xue, Chunbo and Zhou, Zimu and Zhang, Xiaofei and Chen, Lei and Xu, Yi and Xu, Ke and Lv, Weifeng},
+  year = 2022,
+  journal = {Proceedings of the VLDB Endowment},
+  volume = {15},
+  number = {6},
+  pages = {1159--1172},
+  issn = {2150-8097},
+  doi = {10.14778/3514061.3514064},
+  urldate = {2026-01-24}
+}
+
+@article{tong2025hufua,
+  title = {Hu-Fu: efficient and secure spatial queries over data federation},
+  author = {Tong, Yongxin and Zeng, Yuxiang and Song, Yang and Pan, Xuchen and Fan, Zeheng and Xue, Chunbo and Zhou, Zimu and Zhang, Xiaofei and Chen, Lei and Xu, Yi and Xu, Ke and Lv, Weifeng},
+  year = 2025,
+  journal = {The VLDB Journal},
+  volume = {34},
+  number = {2},
+  pages = {19},
+  issn = {1066-8888, 0949-877X},
+  doi = {10.1007/s00778-024-00896-3},
+  urldate = {2026-01-23}
+}
+
@misc{wen2025rsrag,
  title = {RS-RAG: Bridging Remote Sensing Imagery and Comprehensive Knowledge with a Multi-Modal Dataset and Retrieval-Augmented Generation Model},
  author = {Wen, Congcong and Lin, Yiting and Qu, Xiaokang and Li, Nan and Liao, Yong and Lin, Hui and Li, Xiang},
@@ -201,6 +281,30 @@ filename: 1024392702.nh}
  archiveprefix = {arXiv}
 }

+@inproceedings{yang2025joinable,
+  title = {Joinable Search Over Multi-Source Spatial Datasets: Overlap, Coverage, and Efficiency},
+  booktitle = {2025 IEEE 41st International Conference on Data Engineering (ICDE)},
+  author = {Yang, Wenzhe and Wang, Sheng and Chen, Zhiyu and Sun, Yuan and Peng, Zhiyong},
+  year = 2025,
+  pages = {585--598},
+  issn = {2375-026X},
+  doi = {10.1109/ICDE65448.2025.00050},
+  urldate = {2026-01-20}
+}
+
+@article{yin2025list,
+  title = {LIST: learning to index spatio-textual data for embedding based spatial keyword queries},
+  author = {Yin, Ziqi and Feng, Shanshan and Liu, Shang and Cong, Gao and Ong, Yew Soon and Cui, Bin},
+  year = 2025,
+  journal = {The VLDB Journal},
+  volume = {34},
+  number = {3},
+  pages = {33},
+  issn = {1066-8888, 0949-877X},
+  doi = {10.1007/s00778-024-00886-5},
+  urldate = {2026-01-23}
+}
+
@misc{yu2025spatialrag,
  title = {Spatial-RAG: Spatial Retrieval Augmented Generation for Real-World Geospatial Reasoning Questions},
  author = {Yu, Dazhou and Bao, Riyang and Ning, Ruiyu and Peng, Jinghong and Mai, Gengchen and Zhao, Liang},
@@ -214,6 +318,17 @@ filename: 1024392702.nh}
  archiveprefix = {arXiv}
 }

+@inproceedings{zardbani2025updating,
+  title = {Updating an Adaptive Spatial Index},
+  booktitle = {2025 IEEE 41st International Conference on Data Engineering (ICDE)},
+  author = {Zardbani, Fatemeh and Lampropoulos, Konstantinos and Mamoulis, Nikos and Karras, Panagiotis},
+  year = 2025,
+  pages = {1194--1206},
+  issn = {2375-026X},
+  doi = {10.1109/ICDE65448.2025.00094},
+  urldate = {2026-01-20}
+}
+
@article{zhang2025imagerag,
  title = {ImageRAG: Enhancing Ultra High Resolution Remote Sensing Imagery Analysis with ImageRAG},
  author = {Zhang, Zilun and Shen, Haozhan and Zhao, Tiancheng and Guan, Zian and Chen, Bin and Wang, Yuhao and Jia, Xu and Cai, Yuxiang and Shang, Yongheng and Yin, Jianwei},
@@ -229,3 +344,16 @@ filename: 1024392702.nh}
  urldate = {2026-01-19},
  archiveprefix = {arXiv}
 }
+
+@article{zimanyi2020mobilitydb,
+  title = {MobilityDB: A Mobility Database Based on PostgreSQL and PostGIS},
+  author = {Zim{\'a}nyi, Esteban and Sakr, Mahmoud and Lesuisse, Arthur},
+  year = 2020,
+  journal = {ACM Transactions on Database Systems},
+  volume = {45},
+  number = {4},
+  pages = {1--42},
+  issn = {0362-5915, 1557-4644},
+  doi = {10.1145/3406534},
+  urldate = {2026-01-24}
+}
--- a/国内外研究现状/1.时空数据查询与读取.md
+++ b/国内外研究现状/1.时空数据查询与读取.md
@@ -1,3 +1,3 @@
 遥感数据的高效时空查询处理已经得到了广泛的研究，早期的工作主要集中在关系数据库系统中的元数据组织和索引级修剪。传统的方法通常扩展基于树的空间索引，如R-tree，四叉树及其时空变体，将图像足迹与时间属性组织在一起，并且通常在关系后端（例如MySQL和PostgreSQL）上实现。这些方法为中等规模的数据集提供了有效的距离过滤，但随着遥感元数据量的快速增长，它们对平衡树结构的依赖往往导致较高的维护开销和有限的可扩展性。随着数据量和数据摄取速度的不断增加，近年来的系统逐渐转向部署在分布式NoSQL存储上的基于网格的时空索引方案。通过使用GeoHash、GeoSOT或空间填充曲线[@mstgi],[@2024gridmesa]将空间足迹编码为统一的空间网格，并将其与时间标识符结合，这些方法可以实现轻量级索引构建，并在HBase和Elasticsearch等后端具有更好的水平可扩展性。这种基于网格的索引可以通过粗粒度的剪枝有效地减少候选搜索空间，更适合于大规模、持续增长的遥感档案。

-然而，对于遥感工作负载，单靠索引修剪不足以保证端到端的查询效率，因为遥感工作负载中的单个图像通常很大，查询结果需要进一步的像素级处理。为了减少原始I/O的数量，谷歌Earth系统[@gorelick2017google]依赖于平铺和多分辨率金字塔，将图像物理地分成小块。而最近的解决方案则利用COG和基于窗口的I/O来实现对整体映像文件的部分读取。OpenDataCube[@lewis2017australiana]等框架利用这些特性只读取与查询窗口相交的图像区域，从而减少不必要的数据传输。然而，在确定候选图像之后，大多数系统仍然对每个图像执行细粒度的地理空间计算，包括坐标转换和精确的像素窗口推导，当涉及许多图像时，这可能会产生大量开销。
+然而，对于遥感工作负载，单靠索引修剪不足以保证端到端的查询效率，因为遥感工作负载中的单个图像通常很大，查询结果需要进一步的像素级处理。为了减少原始I/O的数量，谷歌Earth系统[@gorelick2017google]依赖于平铺和多分辨率金字塔，将图像物理地分成小块。而最近的解决方案则利用COG和基于窗口的I/O来实现对整体映像文件的部分读取。OpenDataCube[@lewis2017australiana]等框架利用这些特性只读取与查询窗口相交的图像区域，从而减少不必要的数据传输。然而，在确定候选图像之后，大多数系统仍然对每个图像执行细粒度的地理空间计算，包括坐标转换和精确的像素窗口推导，当涉及许多图像时，这可能会产生大量开销。
--- a/国内外研究现状/空间数据库.md
+++ b/国内外研究现状/空间数据库.md
@@ -0,0 +1,23 @@
+过往的空间数据库的主要研究方向包括：
+
+1. 空间数据分区：Shehab 等人最近的一项研究。 [3]涉及SpatialHadoop的增强分区算法。为了为空间数据选择合适的划分方法，在强化学习的背景下进行了一项重要的分析[4]。
+2. 空间索引：Cong 等人。 [30]提出了一种利用倒排文件和 R 树进行 top-k 文本检索的位置感知索引框架。 2018 年，Kraska 等人。 [31]首先引入了学习索引的概念并开发了递归模型索引。 LISA [32] 是一种空间数据的学习索引，它采用机器学习模型来生成数据布局，从而适应不同的数据集。
+3. 空间关键词查询：Ahmed 等人的定性研究。描述了如何查找给定关键字位于前 k 个最常见关键字中的空间区域 [10]。罗等人。通过解决关键字中的印刷错误，将即时空间关键字查询应用于道路网络[11]。
+4. 空间众包：空间众包的核心问题是如何高效地将任务分配给工作人员[33]，[34]。建立了可视化分析系统来呈现实时任务分配并帮助用户分析任务分配的过程[12]。
+
+2025年的论文：
+
+1. 空间关键字查询[@chancskqs] [@yin2025list]。
+2. 空间索引[@zardbani2025updating]。
+3. “Query as Visualization”[@liu2025hivq]：和时空范围查询优化的论文有点像。学习这篇论文的引用。
+
+把旧概念应用于空间数据：
+
+1. 基于数据联合的空间查询[@tong2025hufua], [@tong2022hufu]：**数据联邦**本身也是多源数据，多源数据处理可以引用。
+    1. data isolation problem。
+    2. perform secure queries over a data federation，举了两个例子。
+    3. Nevertheless, directly adapting the state-of-the-art data federation solutions to spatial data can be inefficient，分析原因。
+    4. 目标、边界与方法。
+2. 空间数据库的自然语言接口[@liu2025nalspatial]。
+3. In-Storage Query Processing for Spatio-Temporal Data[@guo2025storagea]：计算卸载。
+4. Joinable Search over Multi-source Spatial Datasets[@yang2025joinablea]：处理**多源数据集成**。