Hello, everyone!
The post will share with you the Elasticsearch
What Is Elasticsearch (ES)?
In the project development, almost every system has a search function. When the data is massive and high-performance search function is required, it is complex to maintain and expand the system search function. What is worse, the full-text retrieval speed, content recommendations, analysis, and statistics aggregation cannot satisfy customers' expectations.
To solve this problem, Elasticsearch is a good choice. It creates full-text indices, which splits the text content into several keywords, and then creates indices based on keywords). By searching for keywords, articles containing the keywords can be found. Elasticsearch isolates data and indices, and divides indices into shards and distributes them to different nodes. Nodes can be expanded to hundreds of nodes, retrieving and processing PB-level structured or unstructured data in a timely manner. Additionally, replica shards can be backed up to ensure data reliability. They work together to greatly improve the retrieval performance. RESTful APIs make the full-text search simplified and efficient.
As Elasticsearch develops rapidly, it provides functions more than search, such as data statistics aggregation and visualization. If you need to locate and analyze millions of documents using keywords, Elasticsearch is the best choice.
Client: Communicates with EsMaster and EsNode instance processes in the Elasticsearch cluster through HTTP or HTTPS, and performs distributed indexing and search operations.
EsMaster: Stores Elasticsearch metadata.
EsNode1 to EsNode9: Store the Elasticsearch index data.
ZooKeeper Cluster: Provides heartbeat mechanism for processes in Elasticsearch clusters.
Features and Advantages
Supports distributed file storage in real time, storing all fields into indices to make the fields retrievable.
Supports distributed real-time search. Indices are divided into shards and each shard can have zero or more replicas. Each node hosts one or more shards, and acts as a coordinator to delegate operations to the correct shards.
Can be expanded to hundreds of servers to process PB-level structured or unstructured data. It can also run on a single PC (this feature has been tested).
Supports the plug-in mechanism, such as analyzer plugin, synchronization plugin, Hadoop plugin, and visualization plugin.
Elasticsearch Usage Constraints
Scenarios Suitable for Elasticsearch
The types of the data to be retrieved are diverse, including structured, semi-structured, and unstructured data.
The search criteria are diversified (for example, too many fields are involved), which cannot be performed by common search engines. Elasticsearch can be used to query full-text retrieval of words, phrases, or words and phrases of different forms.
Data can be retrieved while being written.
Data reads are much more than data writes.
Scenarios Unsuitable for Elasticsearch
Elasticsearch has no concept of transaction, does not support rollback, and cannot restore the deleted data. You are not advised to use Elasticsearch for businesses having special requirements on the transaction of operations.
Remarks
Elasticsearch itself does not have the rights management module.
If no high requirements are required for rights management, a free Search Guard can be used. Search Guard gives full security control over users, roles, and rights, and even controls the rights at the index, type, or host level. Search Guard supports SSL/TLS encryption transmissions of Transport Layer (node-to-node) and REST Layer (HTTP/HTTPS).
If high requirements are required for rights management. You are advised to use a chargeable, official Shield plugin.
Elasticsearch Basic Concept
Please refer to the following link:
https://forum.huawei.com/enterprise/en/elasticsearch-basic-concept/thread/704057-893
That's all, thanks!
