技术领域technical field
本发明属于文件存储技术领域，具体涉及一种基于HDFS小文件写入和读取的方法。The invention belongs to the technical field of file storage, and in particular relates to a method for writing and reading small files based on HDFS.
背景技术Background technique
随着互联网的高速发展，互联网中的数据也急剧的膨胀，为了给用户提供更好的服务，互联网企业就要保存和挖掘这些数据。由此产生了云计算的概念，云计算是当今研究的热门课题, 它很好的解决了大数据的运算与存储的难题，其中云存储作为云计算的衍生也成为了国内外研究的热点，在众多云存储的研究中，Hadoop 的分布式文件系统HDFS作为GoogleFile System 的开源实现已成为业界研究云计算与云存储， 实现云应用提供云服务参考的标准模型。HDFS 可以用于大规模的分布式存储，能够构建一个易扩展、容错性高、高性能的云存储平台，并且它还为用户提供了一组可靠的稳定的接口可以使开发人员根据自己实际的需求进行开发和扩展，Hadoop 目前已得到许多大公司的青睐，它在海量数据的存储与处理上表现优异得到了广泛的应用。With the rapid development of the Internet, the data in the Internet has also expanded rapidly. In order to provide users with better services, Internet companies must save and mine these data. From this came the concept of cloud computing. Cloud computing is a hot research topic today. It solves the problem of computing and storage of big data very well. Among them, cloud storage, as a derivative of cloud computing, has also become a hot research topic at home and abroad. Among many cloud storage studies, Hadoop's distributed file system HDFS, as an open source implementation of Google File System, has become a standard model for the industry to study cloud computing and cloud storage, and to provide cloud services for cloud applications. HDFS can be used for large-scale distributed storage, and can build an easily scalable, high fault-tolerant, high-performance cloud storage platform, and it also provides users with a set of reliable and stable interfaces that enable developers to Hadoop has been favored by many large companies, and it is widely used for its excellent performance in the storage and processing of massive data.
采用HDFS 文件系统能够有效地保持数据的一致性，适合一次写入多次读的场合，该架构可以搭建在任意的计算机上来运行，保障了可扩展性，HDFS 的备份恢复机制与对分配任务的监控机制都保障了分布式存储的可靠性，HDFS 采用的是流式文件读取非常适用于读取海量级数据。然而HDFS 文件系统并非是完美的，它在海量小文件存取上有着一些限制。所以针对HDFS小文件存储问题现在已经有了一些解决方案：The HDFS file system can effectively maintain data consistency and is suitable for the occasion of writing once and reading multiple times. This architecture can be built on any computer to run, ensuring scalability. The monitoring mechanism guarantees the reliability of distributed storage. HDFS uses streaming file reading, which is very suitable for reading massive data. However, the HDFS file system is not perfect, and it has some restrictions on the access of massive small files. So now there are some solutions for HDFS small file storage problem:
（1）将小文件存储到Hbase 中通过文件合并与分解提高文件的存储效率，这种方案的缺点就是随着文件的增多会造成Hbase 大量的合并与分解操作占用大量系统资源严重影响系统的性能，并且Hbase 只支持简单的字符类型，对其它的图片等类型支持不好还需用户单独的处理。(1) Store small files in Hbase to improve file storage efficiency by merging and decomposing files. The disadvantage of this solution is that as the number of files increases, a large number of merging and decomposing operations in Hbase will take up a lot of system resources and seriously affect system performance. , and Hbase only supports simple character types, and it needs to be handled separately by the user if it does not support other types of pictures.
（2）还有就是利用Hadoop 提供的归档工具Hadoop Archives (简称HAR Files)对小文件进行归档打包，这种方式虽然能够有效的降低大量小文件对NameNode 的内存消耗但是当用户需要访问时需要超找两次索引才能找到文件效率不高，同时还需要管理员维护运行命令进行归档操作不适合构建基于互联网的云存储平台。(2) There is also the use of Hadoop Archives (referred to as HAR Files), an archiving tool provided by Hadoop, to archive and package small files. Although this method can effectively reduce the memory consumption of a large number of small files on the NameNode, it needs to be overtime when users need to access It is not efficient to search for the index twice to find the file. At the same time, administrators need to maintain and run commands to perform archiving operations. It is not suitable for building an Internet-based cloud storage platform.
（3）还有一种文件合并的方案是利用Sequence File 来进行文件的合并，Sequence File 是用来存储二进制key-value 形式的文件，通常利用Sequence File 来存储小文件时将文件名存入到key 中文件内容存储到value中，这种方式最大的缺点就是由于其中的键值是未排序的文件随机读取效率较低，需要遍历整个文件才能进行读取，并且这种方式不支持文件追加操作，因此合并之前的小文件要在服务器进行缓存，这样文件的安全性就不能得到保障。(3) Another file merging scheme is to use Sequence File to merge files. Sequence File is used to store files in the form of binary key-value. Usually, when Sequence File is used to store small files, the file name is stored in the key. The content of the file is stored in the value. The biggest disadvantage of this method is that the random reading efficiency of the unsorted file is low, and the entire file needs to be traversed to read, and this method does not support file append operations. , so the small files before merging should be cached on the server, so the security of the files cannot be guaranteed.
发明内容Contents of the invention
本发明的目的在于，针对上述现有技术存在的缺陷，提供设计一种基于HDFS小文件写入和读取的方法，以解决上述技术问题。The object of the present invention is to provide and design a method for writing and reading small files based on HDFS to solve the above-mentioned technical problems.
为了实现上述目的，本发明的技术方案为：In order to achieve the above object, the technical solution of the present invention is:
一种基于HDFS小文件写入和读取的方法，其特征在于，包括以下两部分：（1）文件的写入A method for writing and reading small files based on HDFS, characterized in that it includes the following two parts: (1) writing of files
Sw1：发送上传文件请求；Sw1: Send a file upload request;
Sw2: 获取待上传文件的大小，并将待上传文件的大小与设置的阈值进行比较，若待上传文件的大小小于设置的阈值，则认为是小文件，判断用户文件是否存在，若不存在，执行步骤Sw3；若存在，跳转至步骤Sw4，Sw2: Obtain the size of the file to be uploaded, and compare the size of the file to be uploaded with the set threshold. If the size of the file to be uploaded is smaller than the set threshold, it will be considered as a small file, and determine whether the user file exists. If not, Execute step Sw3; if exists, jump to step Sw4,
Sw3：在HDFS集群中建立新的用户文件并且用户文件的名称与用户ID一一对应，则以追加写的方式将小文件合并到用户文件中，执行步骤Sw5；Sw3: Create a new user file in the HDFS cluster and the name of the user file is in one-to-one correspondence with the user ID, then merge the small file into the user file in the form of additional writing, and execute step Sw5;
Sw4：判断小文件大小是否大于用户文件剩余空间的大小，若是，跳转步骤Sw3，若否，则将文件以追加写的方式追加到用户文件中；Sw4: Determine whether the size of the small file is greater than the size of the remaining space of the user file, if so, jump to step Sw3, if not, append the file to the user file in the form of append writing;
Sw5：将小文件的元数据信息存入到关系型数据库中，并建立查询索引。Sw5: Store the metadata information of the small file into the relational database, and establish a query index.
（2）文件的读取(2) Reading of files
Sr1：发送读取文件请求；Sr1: Send a read file request;
Sr2：获取文件的大小，并将文件的大小与设置的阈值进行比较，若文件的大小小于设置的阈值时，跳转至Sr4，当文件大于设定的阈值时执行步骤Sr3;Sr2: Obtain the size of the file, and compare the size of the file with the set threshold, if the size of the file is smaller than the set threshold, jump to Sr4, and execute step Sr3 when the file is larger than the set threshold;
Sr3：如果文件大于设定的阈值并且小于一个文件块的大小，将其请求转发到数据节点上，通过函数得到文件块的相关信息，直接由数据节点来向客户端传送数据；Sr3: If the file is larger than the set threshold and smaller than the size of a file block, the request is forwarded to the data node, and the relevant information of the file block is obtained through the function, and the data node directly transmits the data to the client;
Sr4：读取文件的偏移值与大小，通过函数定位到小文件所在用户文件中的位置，根据元数据信息以流式的方式读取小文件；Sr4: Read the offset value and size of the file, locate the location of the small file in the user file through the function, and read the small file in a streaming manner according to the metadata information;
进一步的，步骤Sw2中，待上传文件的大小大于设置的阈值，则认为是大文件，直接由通用文件模块上传。Further, in step Sw2, if the size of the file to be uploaded is greater than the set threshold, it is considered as a large file and is directly uploaded by the general file module.
步骤Sw5中，存在关联性的小文件的元数据信息存放于同一目录下。In step Sw5, the metadata information of associated small files is stored in the same directory.
步骤Sr3：文件大于一个文件块的大小时，直接读取文件。Step Sr3: When the file is larger than the size of one file block, read the file directly.
所述元数据信息包括文件名称、大小、位置、属性、创建时间、修改时间、偏移值、长度length。The metadata information includes file name, size, location, attribute, creation time, modification time, offset value, and length.
所述文件块的相关信息包括，数据节点的名称、端口号、该文件在块中的偏移量。The relevant information of the file block includes the name of the data node, the port number, and the offset of the file in the block.
用户文件为一个文件块大小的文件。A user file is a file with the size of a file block.
所述文件块大小默认为64M。The file block size is 64M by default.
所述阈值默认为1M。The threshold is 1M by default.
本发明的有益效果在于：The beneficial effects of the present invention are:
本文发明提出了关系数据库的小文件合并算法，为每个用户都建立一个用户文件，每个用户文件都不同，用文件追加操作以流式的形式将小文件合并到这个用户文件中，并且还利用关系数据库为小文件建立索引记录小文件的偏移信息，提高了小文件的读取效率，同时当访问小于一个块的文件时，将其请求转发到数据节点上处理减轻主服务器的压力提高文件传送效率；将小文件合并为一个文件存储NameNode 内存并不会随着文件的增多而改变，从而降低了NameNode 内存的消耗，提升了整个系统的性能；小文件合并有效的解决文件格式的限制, 还能将用户小文件随时追加到用户文件中避免先将文件缓存到服务器达到一定大小后再上传产生的文件安全性问题, 最后通过为小文件建立索引信息能够有效的提高文件的检索效率。This paper proposes a small file merging algorithm for relational databases. A user file is created for each user, and each user file is different. The file append operation is used to merge small files into this user file in a streaming form, and also Use the relational database to build indexes for small files and record the offset information of small files, which improves the reading efficiency of small files. At the same time, when accessing files smaller than one block, the request is forwarded to the data node for processing, reducing the pressure on the main server. File transfer efficiency; merging small files into one file to store NameNode memory will not change with the increase of files, thereby reducing the consumption of NameNode memory and improving the performance of the entire system; merging small files can effectively solve the limitation of file format , can also add small user files to user files at any time to avoid file security issues caused by first caching files to the server to a certain size and then uploading them. Finally, indexing information for small files can effectively improve file retrieval efficiency.
此外，本发明设计原理可靠，结构简单，具有非常广泛的应用前景。In addition, the design principle of the present invention is reliable, the structure is simple, and has very wide application prospects.
由此可见，本发明与现有技术相比，具有突出的实质性特点和显著地进步，其实施的有益效果也是显而易见的。It can be seen that, compared with the prior art, the present invention has outstanding substantive features and remarkable progress, and the beneficial effects of its implementation are also obvious.
附图说明Description of drawings
图1为一种基于HDFS小文件写入方法流程图。Fig. 1 is a flow chart of a method for writing small files based on HDFS.
图2为一种基于HDFS小文件读取方法流程图。FIG. 2 is a flowchart of a method for reading small files based on HDFS.
具体实施方式detailed description
下面结合附图并通过具体实施例对本发明进行详细阐述，以下实施例是对本发明的解释，而本发明并不局限于以下实施方式。The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are explanations of the present invention, but the present invention is not limited to the following embodiments.
本实施例提供的一种基于HDFS小文件写入和读取的方法，如图1所示，当用户写入文件时，首先向NameNode发送上传文件请求，之后进行大小判断，如果是大文件则直接由通用文件模块上传到HDFS集群中，如果小文件，则交给小文件处理模块，如果文件大小小于设置的阈值1M，我们就将文件以追加写的方式追加到HDFS集群中该用户的用户文件中，写入完成后我们要将文件的大小与偏移值记录到关系型数据库中，并建立索引，存在关联性的小文件的元数据信息存放于同一目录下，若小文件大小大于用户文件剩余空间的大小，在HDFS集群建立新的用户文件并且用户文件的名称与用户ID一一对应，则以追加写的方式将小文件合并到新建的用户文件中。HDFS文件系统中默认是不支持追加写操作的，因此需要进行配置，修改Namenode服务器下的HDFS-site.xml文件，将dfs.support.append的值设为true。A method for writing and reading small files based on HDFS provided by this embodiment, as shown in Figure 1, when a user writes a file, first sends a file upload request to the NameNode, and then judges the size, if it is a large file, then Upload directly to the HDFS cluster by the general file module. If the file is small, it will be handed over to the small file processing module. If the file size is less than the set threshold of 1M, we will append the file to the user of the user in the HDFS cluster in the form of append writing. In the file, after the writing is completed, we need to record the size and offset value of the file in the relational database and build an index. The metadata information of small related files is stored in the same directory. If the size of the small file is larger than the user The size of the remaining space of the file, if a new user file is created in the HDFS cluster and the name of the user file corresponds to the user ID one by one, the small file will be merged into the newly created user file by means of append writing. By default, the HDFS file system does not support additional write operations, so it needs to be configured, modify the HDFS-site.xml file under the Namenode server, and set the value of dfs.support.append to true.
如图2所示，当用户向NameNode发起读取文件请求时，首先要对文件大小进行判断，当文件小于设置的阈值1M时，小文件处理模块就从关系型的数据库中读取文件的偏移值与大小，然后通过seek()函数定位到小文件所在用户文件中的位置，最后根据length长度以流式的方式从HDFS集群中读取小文件。当文件大于设定的阈值1M时就进行进一步的判断，如果文件大于一个文件块的大小(默认为64M)时，直接从HDFS集群中读取文件。如果文件小于一个块的大小，小文件处理模块，则通过getFileBlockLocation()函数得到文件块的相关信息，其中包括数据节点的名称与端口号以及该文件在块中的偏移量等信息，直接由数据节点来向客户端传送数据，提高了文件传送效率，减轻NameNode的压力。As shown in Figure 2, when the user initiates a file read request to the NameNode, the file size must first be judged. When the file is smaller than the set threshold of 1M, the small file processing module reads the partial file size from the relational database. Move the value and size, and then use the seek() function to locate the location of the small file in the user file, and finally read the small file from the HDFS cluster in a streaming manner according to the length. When the file is larger than the set threshold of 1M, further judgment is made. If the file is larger than the size of a file block (64M by default), the file is directly read from the HDFS cluster. If the file is smaller than the size of a block, the small file processing module will obtain the relevant information of the file block through the getFileBlockLocation() function, including the name and port number of the data node and the offset of the file in the block, etc., directly by The data node is used to transmit data to the client, which improves the efficiency of file transmission and reduces the pressure on the NameNode.
以上公开的仅为本发明的优选实施方式，但本发明并非局限于此，任何本领域的技术人员能思之的没有创造性的变化，以及在不脱离本发明原理前提下所作的若干改进和润饰，都应落在本发明的保护范围内。The above disclosure is only a preferred embodiment of the present invention, but the present invention is not limited thereto, any non-creative changes that those skilled in the art can think of, and some improvements and modifications made without departing from the principle of the present invention , should fall within the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710100365.5ACN106909651A (en) | 2017-02-23 | 2017-02-23 | A kind of method for being write based on HDFS small documents and being read |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710100365.5ACN106909651A (en) | 2017-02-23 | 2017-02-23 | A kind of method for being write based on HDFS small documents and being read |
| Publication Number | Publication Date |
|---|---|
| CN106909651Atrue CN106909651A (en) | 2017-06-30 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710100365.5APendingCN106909651A (en) | 2017-02-23 | 2017-02-23 | A kind of method for being write based on HDFS small documents and being read |
| Country | Link |
|---|---|
| CN (1) | CN106909651A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107506447A (en)* | 2017-08-25 | 2017-12-22 | 郑州云海信息技术有限公司 | A kind of small documents reading/writing method and system based on local file system |
| CN107729432A (en)* | 2017-09-29 | 2018-02-23 | 浪潮软件股份有限公司 | A kind of storage of distributed small documents, read method, device and access system |
| CN108234594A (en)* | 2017-11-28 | 2018-06-29 | 北京市商汤科技开发有限公司 | File memory method and device, electronic equipment, program and medium |
| CN108595567A (en)* | 2018-04-13 | 2018-09-28 | 郑州云海信息技术有限公司 | A kind of merging method of small documents, device, equipment and readable storage medium storing program for executing |
| CN108717457A (en)* | 2018-05-23 | 2018-10-30 | 苏州易康萌思网络科技有限公司 | A kind of e-commerce platform big data processing method and system |
| CN108932287A (en)* | 2018-05-22 | 2018-12-04 | 广东技术师范学院 | A kind of mass small documents wiring method based on Hadoop |
| CN110069451A (en)* | 2019-03-28 | 2019-07-30 | 浪潮卓数大数据产业发展有限公司 | A kind of method and device of HDFS storage small documents |
| CN110413588A (en)* | 2019-07-30 | 2019-11-05 | 中国工商银行股份有限公司 | Distributed objects storage method, device, computer equipment and storage medium |
| CN110515920A (en)* | 2019-08-30 | 2019-11-29 | 北京浪潮数据技术有限公司 | A kind of mass small documents access method and system based on Hadoop |
| CN113905037A (en)* | 2021-06-18 | 2022-01-07 | 武汉理工数字传播工程有限公司 | File transmission management method, device, equipment and storage medium |
| CN114968939A (en)* | 2022-05-31 | 2022-08-30 | 济南浪潮数据技术有限公司 | A file merging method, device and computer-readable storage medium |
| CN116737668A (en)* | 2023-05-31 | 2023-09-12 | 北京焱融科技有限公司 | A method to improve file append writing performance under distributed file systems |
| CN116756093A (en)* | 2023-08-17 | 2023-09-15 | 天津神舟通用数据技术有限公司 | Large object storage and query method, device, equipment and medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102332027A (en)* | 2011-10-15 | 2012-01-25 | 西安交通大学 | A method for associative storage of massive non-independent small files based on Hadoop |
| CN103577123A (en)* | 2013-11-12 | 2014-02-12 | 河海大学 | Small file optimization storage method based on HDFS |
| CN103856567A (en)* | 2014-03-26 | 2014-06-11 | 西安电子科技大学 | Small file storage method based on Hadoop distributed file system |
| US20150379024A1 (en)* | 2014-06-27 | 2015-12-31 | International Business Machines Corporation | File storage processing in hdfs |
| CN105631010A (en)* | 2015-12-29 | 2016-06-01 | 成都康赛信息技术有限公司 | Optimization method based on HDFS small file storage |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102332027A (en)* | 2011-10-15 | 2012-01-25 | 西安交通大学 | A method for associative storage of massive non-independent small files based on Hadoop |
| CN103577123A (en)* | 2013-11-12 | 2014-02-12 | 河海大学 | Small file optimization storage method based on HDFS |
| CN103856567A (en)* | 2014-03-26 | 2014-06-11 | 西安电子科技大学 | Small file storage method based on Hadoop distributed file system |
| US20150379024A1 (en)* | 2014-06-27 | 2015-12-31 | International Business Machines Corporation | File storage processing in hdfs |
| CN105631010A (en)* | 2015-12-29 | 2016-06-01 | 成都康赛信息技术有限公司 | Optimization method based on HDFS small file storage |
| Title |
|---|
| 张海 等: "基于HDFS的小文件存储与读取优化策略", 《计算机系统应用》* |
| 张海: "基于HDFS分布式存储技术研究与优化", 《中国优秀硕士学位论文全文数据库 信息科技辑》* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107506447A (en)* | 2017-08-25 | 2017-12-22 | 郑州云海信息技术有限公司 | A kind of small documents reading/writing method and system based on local file system |
| CN107729432A (en)* | 2017-09-29 | 2018-02-23 | 浪潮软件股份有限公司 | A kind of storage of distributed small documents, read method, device and access system |
| CN108234594A (en)* | 2017-11-28 | 2018-06-29 | 北京市商汤科技开发有限公司 | File memory method and device, electronic equipment, program and medium |
| CN108595567A (en)* | 2018-04-13 | 2018-09-28 | 郑州云海信息技术有限公司 | A kind of merging method of small documents, device, equipment and readable storage medium storing program for executing |
| CN108932287B (en)* | 2018-05-22 | 2019-11-29 | 广东技术师范大学 | A kind of mass small documents wiring method based on Hadoop |
| CN108932287A (en)* | 2018-05-22 | 2018-12-04 | 广东技术师范学院 | A kind of mass small documents wiring method based on Hadoop |
| CN108717457A (en)* | 2018-05-23 | 2018-10-30 | 苏州易康萌思网络科技有限公司 | A kind of e-commerce platform big data processing method and system |
| CN110069451A (en)* | 2019-03-28 | 2019-07-30 | 浪潮卓数大数据产业发展有限公司 | A kind of method and device of HDFS storage small documents |
| CN110413588A (en)* | 2019-07-30 | 2019-11-05 | 中国工商银行股份有限公司 | Distributed objects storage method, device, computer equipment and storage medium |
| CN110515920A (en)* | 2019-08-30 | 2019-11-29 | 北京浪潮数据技术有限公司 | A kind of mass small documents access method and system based on Hadoop |
| CN113905037A (en)* | 2021-06-18 | 2022-01-07 | 武汉理工数字传播工程有限公司 | File transmission management method, device, equipment and storage medium |
| CN114968939A (en)* | 2022-05-31 | 2022-08-30 | 济南浪潮数据技术有限公司 | A file merging method, device and computer-readable storage medium |
| CN116737668A (en)* | 2023-05-31 | 2023-09-12 | 北京焱融科技有限公司 | A method to improve file append writing performance under distributed file systems |
| CN116756093A (en)* | 2023-08-17 | 2023-09-15 | 天津神舟通用数据技术有限公司 | Large object storage and query method, device, equipment and medium |
| CN116756093B (en)* | 2023-08-17 | 2023-11-03 | 天津神舟通用数据技术有限公司 | Large object storage and query method, device, equipment and medium |
| Publication | Publication Date | Title |
|---|---|---|
| CN106909651A (en) | A kind of method for being write based on HDFS small documents and being read | |
| CN102662992B (en) | Method and device for storing and accessing massive small files | |
| CN102014158B (en) | Cloud storage service client high-efficiency fine-granularity data caching system and method | |
| US8819335B1 (en) | System and method for executing map-reduce tasks in a storage device | |
| CN102609360B (en) | Data processing method, data processing device and data processing system | |
| CN104731921B (en) | Storage and processing method of the Hadoop distributed file systems for log type small documents | |
| CN102364474B (en) | Metadata storage system for cluster file system and metadata management method | |
| CN103577123B (en) | A kind of small documents optimization based on HDFS stores method | |
| CN103020315B (en) | A kind of mass small documents storage means based on master-salve distributed file system | |
| CN100483420C (en) | Fine grit document and catalogs version management method based on snapshot | |
| CN114281762B (en) | A log storage acceleration method, device, device and medium | |
| CN105183839A (en) | Hadoop-based storage optimizing method for small file hierachical indexing | |
| CN110647497A (en) | A high-performance file storage and management system based on HDFS | |
| CN103595797B (en) | Caching method for distributed storage system | |
| CN104424219B (en) | A kind of management method and device of data file | |
| CN106021381A (en) | Data access/storage method and device for cloud storage service system | |
| CN106775446A (en) | Based on the distributed file system small documents access method that solid state hard disc accelerates | |
| CN103139300A (en) | Virtual machine image management optimization method based on data de-duplication | |
| CN103473337A (en) | Massive catalogs and files oriented processing method in distributed type storage system | |
| WO2022262381A1 (en) | Data compression method and apparatus | |
| CN109933564A (en) | File system management method, device, terminal and medium for fast rollback based on linked list and N-ary tree structure | |
| CN115098466A (en) | Metadata management method and device, storage node and readable storage medium | |
| US20250278410A1 (en) | Hybrid transactional and analytical processing architecture for optimization of real-time analytical querying | |
| US12007983B2 (en) | Optimization of application of transactional information for a hybrid transactional and analytical processing architecture | |
| CN103246716B (en) | Based on object copies efficient management and the system of object cluster file system |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20170630 |