Elasticsearch introduction of the beginning of the Ling (a)

Elasticsearch is also a very good open-source full-Text search framework, based on the famous Apache Lucene, highly encapsulated in a richer, easier to use API, as well as Apache SOLR, Provides a very powerful distributed cluster function!

Do not understand Elasticsearch (hereinafter referred to as ES) is a friend of what to do, you can click on the link on the encyclopedia to get a general understanding of the next.


The following into the topic, this chapter on the introduction of the first macro es some of the concepts and characteristics, so that we can have an understanding of the whole ES, follow-up, the scattered fairy will write some how to install, deployment, tuning, use, clustering, etc.

Es, like LUCENE,SOLR, are non-modal, column-based storage formats, which are the same as most NoSQL databases, very flexible, and below we look at the relational database mapping into ES, the corresponding noun relationship:

 

1, a restful search engine based on  
2, highly available  
3, each index is a shard  based on the configuration punch;
4, each shard can have one or more replicas  
5, the read operation automatically loads to any one copy  
6, Multi-tenant and multi-type  
7, supports more than one index (similar database)  
8, one index supports multiple types (similar to tables)  
9, Flexible index configuration, number of shards, storage type  
10, various set API interfaces  
11, RESTful api  based on HTTP;
12, perfect support native Java api 
13, all API operation, support automatic routing and forwarding  
14, storage mode   for document,
15, Dynamic Data type configuration without prior definition  
16,schema can customize the operation of each type  
17, long-term high-reliability, asynchronous write support  ,
18, near real-time search results (note is near real-time, not real-time)  
19, based on Apache luecene 
20, each shard is a fully functional Lucene index  
21, all Lucene functions can be implemented in ES with simple configuration and plug-in  
22, ensure operation consistency  
23, a single doc (equivalent to a row of data in a relational database) level operation, support a (atomic), C (consistent), I (Isolated), D (persistent) characteristics  
24, fully based on Lucene Open Source enterprise-wide full-text retrieval framework  

Es is a highly reliable, distributed search engine, Each large index is split into multiple shards, each supporting one or more replicas, and by default, a large index is split into 5 shards, each shard a copy (5/1), and of course, we can configure our topology to suit our own situation, Including 1 shard,10 copies (1/10), support read more write less scenes, or 20 Shard, a backup, support write more, read less scenes, of course, when we search, will be similar to a map,reduce process, multiple map retrieval, Finally, the results are merged in a reduce.  

In order to support the distributed nature of ES, each shard is deployed on one or more node (machine nodes), supports start-up and shutdown, as long as it ensures that we write the correct HTTP port, the system will continue to accept the request, for us to query the last indexed data  

Job search engineer  
Some requirements:  

1, skilled use of lucene,solr 
2, there are e-commerce search experience priority  
3, more than 3 years of Java development experience  
4, above the level of education can  

The main company is to do cross-border business e-commerce platform  
Address   near Sanyuan Bridge;
Welfare: Words Supplement, rice supplement    
salary between 15k-20k  

Interested parties please leave a message to contact the scattered fairy, Help recommend!  

> Welcome to the public: I'm an engineer (Woshigcs) and more about the Siege division, here.  
QR code scan:  

 

This article is from the "7936494" blog, please be sure to keep this source http://7946494.blog.51cto.com/7936494/1596356

Elasticsearch introduction of the beginning of the Ling (a)