ElasticSearch is the search engine of choice for companies that use big data: content search, data analysis, queries, and suggestions, especially because of its powerful, powerful, and flexible capabilities. Examples of large companies using ElasticSearch are LinkedIn, Stack Overflow, Foursquare, GitHub, and Amazon.

To be able to store and analyze big data (big data) correctly, you must first be fully aware of your needs. If you need a full-text search between large data chunks, Elasticsearch may be the right choice and if you want to learn more about the elastic search then take elasticsearch training.

Elasticsearch searches the index instead of searching the text directly and produces results very quickly. In addition, you can perform statistical analysis and evaluate queries.

• It's open-source. It is based on Java.
• Based on Apache Lucene (the concept of sharing and replica is also valid in ES).
• The data storage format is document-oriented, not relational.
• You can make automatic assignments according to the data type.
• It is a cluster structure, and the cluster structure is very simple.
• Can operate in a distributed and scalable structure.
• Enables analysis of real-time data.
• Concepts include types, fields, documents, fields, notes, full-text search, indexes, and more.
• It is available in all programming languages ​​as it is serviced via the Restfull API.
• Can be used with Kibana, which can monitor Elasticsearch and Logstash tools to host logs.
• Already offers a high level of accessibility.
• We promise quick installation and easy configuration. Getting data into
• Elasticsearch is very easy.
• There is a River-like structure for data transfer.
• Index the document as JSON.

Elasticsearch Infrastructure

Elasticsearch is a completely Java open source project with a distributed architecture. The search engine is based on the search engine Apache Lucene project. The Lucene project primarily provides plain text indexing and searching. This is called unstructured. Elasticsearch, on the other hand, provides what's called a structured data structure on the Lucene infrastructure for indexing and searching for specific documents. 

How does Elasticsearch perform a search on text?

Elasticsearch shows in which document (row) the word appears when the data is saved. Then, if you search for a word instead of searching for all the data, the results will be immediately found in the index list you created earlier. For example, if you search for a word holder in the following document, you can quickly get 1, 4, 5 results from the index list.

Benefits of Elasticsearch Framework:

• SPEED

First of all,  the most important feature of the Elasticsearch Framework is very fast. In principle, you can include any data that can be included in a relational database system, and you can query this data much faster than a database system. To provide this speed, we use a special index structure that can contain numbers, geographic values, dates, and text.

Basically, you can assign data in JSON format from Elasticsearch with trillions of rows and access it in milliseconds with proper configuration and an easy-to-use API.

• Scalability

Elasticsearch can be run on a laptop or server as needed. The cluster installation is fully automatic by the Elasticsearch infrastructure. In addition, communicating with elasticsearch installed on your laptop, and with an ElasticSearch cluster consisting of thousands of servers, is just as easy. Elasticsearch also automatically determines how to keep your data and indexes.

• Ease of use

Elasticsearch is very easy to use thanks to the REST API provided. In addition, it can be used in many programming languages ​​such as Java, C #, Python, Javascript, PHP, Ruby, and special libraries have been created for each language.

Elasticsearch Components and Core Principles:

• Index

The index concept is used in Elasticsearch instead of the traditional relational database. Elasticsearch clusters can contain multiple indexes (databases). Type ES uses the concept of "types" for tables in relational databases. The index can contain multiple types (tables).

• Document

In Elasticsearch, it is represented as a line document in a relational database. Each type has multiple documents.

• Fields

Classic database columns are qualified as Elasticsearch fields. Each document has multiple fields.

• Full-Text Search

This is the name for quick access to the results found by searching for any keyword in the text document and documents that match the keyword from any source. Suppose you search wikipedia.com for the name of a scientist and read an article about it. Instead of searching all registered articles to find an article, enter Nikola Tesla in the Wikipedia search bar and use the structure to text the name of Nikola Tesla under the previously indexed data. You can go to the index that contains and enter it quickly. All things you can learn how things work in elasticsearch from best elasticsearch course.

• Each record added to the index

Elasticsearch is organized as a JSON document. For every word in a document, there is an indexing system that stores information about which document it contains. You can think of it as a kind of database. Similar to the order of the data in the database, Elasticsearch indexes are organized in JSON format.

• Mapping

When indexing data, you need to indicate the type of data. That is, when you index a word, it is the process of defining the data type (string, integer, Boolean) of the word.

• Near real-time

Elasticsearch runs almost real-time. The reason it is called "near real-time" rather than real-time is that it indexes documents with slightly different latencies than real-time. (This period is usually a delay of 1 second.)

• Clusters

Can be referred to as a cluster, or a  collection of nodes consisting of multiple nodes that hold all data and perform all indexing and search functions. increase. The cluster is defined by default with the unique name "elasticsearch". This name can be changed as an option. Naming is important because nodes can only be part of a cluster. If the node is configured to join the cluster with a cluster name, the definitions of these names must be correct.