Shulou(Shulou.com)06/01 Report--

In this issue, Xiaobian will bring you the basic concept of Elasticsearch. The article is rich in content and analyzed and described from a professional perspective. After reading this article, I hope you can gain something.

ES is a highly scalable, open-source, full-text search engine that provides near-real-time indexing, search, and analysis capabilities. Shay Banon is the founder of ES, a Java implementation based on Apache Lucene. Lucene is very complex, and ES shields Lucene's complexity through RESTful APIs, providing a convenient application interface.

application scenarios

It provides a powerful search function, which can be similar to Baidu, Google and other searches.

Log or transaction data can be searched to analyze business trends, collect logs, analyze system bottlenecks or operational developments, etc.

It can provide early warning function (continuous query analysis of a certain data, if it exceeds a certain value, it will be warned)

Analyze business information and easily locate key information in millions of big data

ES Typical Applications

Wikipedia

Stack Overflow

GitHub

Docker

basic concepts

Cluster: A Cluster is an integration of a set of servers that collaboratively store, index, retrieve, and parse data. Cluster has a unique ID name, you only need to specify the cluster ID name (default is elasticsearch), when starting, all clusters with this name will be added to a cluster by default, the election of master nodes and node management are automatically completed (cluster deployment reference article ElasticSearch deployment).

Node: A Node is a single server node that participates in a Cluster, has a unique identification name, and can be added to a specified Cluster.

Index: Index is a collection of documents, a collection of data documents with the same business characteristics (not the same data structure), equivalent to a database in a traditional database. Indexing, searching, and analyzing ES data is done based on the index. Any number of indexes can be created in a Cluster.

Type: Type is the data in Index, used to identify a collection of different document field information, equivalent to a table in a traditional database. Versions after 2.0 directly check insertions, prohibiting conflicts between field types of different types under an index.

Document: Document is the basic storage unit in which ES data can be indexed. It needs to be stored in Type, which is equivalent to the row record of traditional database. It is represented by json.

Shard/Shard: Shard is the ES basic storage container and is an instance of Lucene, so it has full storage search analysis capabilities. Shard is the foundation for ES scalability and reliability. ES creates 5 shards by default when creating an index, and this number can be modified. The number of shards can only be specified at the time of index creation and cannot be modified later.

If an index has a large amount of data, it will cause bottlenecks in hardware hard disk and search speed. If divided into multiple slices, the slices can share the stress.

Fragmentation allows users to scale and split horizontally

Fragmentation allows distributed operations, which can improve the efficiency of search and other operations

Replication/Backup: Copying a shard completes the backup of the shard. The number of backups can be defined dynamically.

When a shard fails or goes offline, backup shards can replace work, improving high availability.

Backup shards can also perform search operations, taking the burden off the search.

NRT(Near Real Time): ES is not a standard database, unlike MongoDB, which focuses on searching stored data. So notice that it's not read and write in real time, which means that the data that has just been stored cannot be queried immediately. By default, the shard is updated every second, so there is a delay of 1s.

The above is what the basic concept of Elasticsearch shared by Xiaobian is. If there is a similar doubt, please refer to the above analysis for understanding. If you want to know more about it, please pay attention to the industry information channel.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.