编程知识 cdmana.com

Elasticsearch, you deserve it! ——Panoramic review of the application of elasticsearch

1、 Why attend the cloud habitat Conference ?

This year it's online , Attending the conference is actually watching live or replaying videos .

I mentioned on many occasions that my idol is Yu Minhong , There was such a passage in his speech at college many years ago :“ My chat with Ma Yun , Do you want to hear ? Of course I want to hear , Because we are talking about things we can't see in books ”.

Yes , The big stars gathered at the yunqi conference , As Elastic lovers , To attend the conference is to see that you can't see 、 What you can't learn .

In the four dimensions of cognition “ I don't know I don't know ” Most terrifying .

There was such a discussion in wechat group before :

[ In fact, it can be done a lot es In the beginning, it was more text related search . But in fact, information is more graphical now 、 Video 、 Other media types . In fact, it's very important . It's raw data embeding Generating vectors , Storage vector , Search for similarity .Elasticsearch There is still a lot of room for what needs to be done in this area ]

It's caused people to talk about Elasticsearch A discussion of the future direction of development .

This cloud habitat Conference Elasticsearch Scenario application gives the answer .

We from :Elastic Stack Three solutions 、 from Zero To Hero Learning path 、 Alibaba cloud Elasticsearch Bull force place to do next double dish .

2、Elastic Stack Three solutions

2.1 “3 + 1“ Strategic scenarios

“3 + 1” It means in Elastic Stack On the basis of Elastic Enterprise search 、Elastic Full observation 、Elastic Security Three core business scenarios .

  • 1: It stands for ——Elastic Stack.

  • 3: It stands for —— Search for 、 Observability 、 Security .

1) Search for —— Don't introduce too much , Traditional search scenarios , such as : Web search 、APP Search for etc. .

2) Observability —— It's around enterprise infrastructure data , journal 、 indicators 、APM, The trinity .

ps: I always thought : Observability and Full observation Chinese is not easy to understand , Not even in Cihai . The essence is : Data visualization means .

English means :Elastic Observability. This translation of the latest official document , It's relatively easy to understand .

3) Safety is Elastic The focus is on one direction of the force , This piece of Elastic It's also a subverter of an industry .

Elastic It is the first manufacturer in the industry to provide open source and free terminal security solutions , Recently, it also released its own open source detection rule library , to open up 、 Transparency and community collaboration will be the only choice for dealing with large-scale security threats in the future , Enterprises can use Elastic Of SIEM And terminal protection scheme , Build your own SOC The team , Protecting the digital assets of enterprises .

In a word :Elastic It has been changed from the original Elasticsearch Search tools have been transformed into Elastic Stack Solution providers .

2.2. How to achieve full data visibility ?

2.2.1 Make data collection easier

  • Elastic Agent

  • solve :beats many , It's not easy to manage  

  • More shielding agent Complexity issues , Reduce installation complexity

  • Fleet

  • Clusters manage a lot of Agent

  • ingest Manager

  • UI management , Docking various data sources

  • Better out of the box

  • Data stream

  • Better manage the life cycle of data for different datasets

2.2.2 Asynchronous search , Carrying more data

For big data offline non real time business scenarios , There is no requirement for retrieval response time , But it's very useful to be concerned about the return of results .

2.2.3 Heap memory upgrade

To improve search 、 Polymerization properties ,Elasticsearch I've been iterating fast to upgrade myself .

2.2.4 Better visualization

  • Chart drill down 、 Different dashboad Jump

Anyone who has done big data analysis knows , It's painful not to be able to run in , Also need help. :echart Wait for yourself .

  • kibana Lens Continuous improvement Support pie charts 、 Cycle graph 、 Tree structure

Chart & The analysis is constantly updated

  • New alarm system Integration solutions , It's easier to configure .

  • APM Sevice Map Full field automatic monitoring

2.2.5 Better analysis and presentation

In a word :Elasticsearch To change the life of data analysis !

Traditional data analysts rely on Elastic stack Can be more efficient to implement the analysis work , It's just “ be a tiger with wings added ”.

even to the extent that : Some developers 、 Operation and maintenance personnel can realize their own multi-dimensional data analysis .

3、 from Zero To Hero Elastic Stack Learning path

Mr. Li Meng has a fluent style of writing 、 Comment on spicy , Wrote a lot of beautiful Elasticsearch Practical blog post .

In this version, Mr. Li Meng talked about his own ideas , The summary is as follows .

3.1 Systematic learning

Abandon assembly learning 、 Fragmented learning thinking .

You bet , I have a great sense of accomplishment at the moment of problem solving , But there's no systematic thinking , It's easy to come up with either “ limited outlook ”、 or “ take unnecessary pains to study an insignificant problem ”.

3.2 How to improve ?

1、 Be very correct in your cognition

2、 Learn with a broad mind Elasticsearch,ES Deficiencies can be combined with other programs to improve .

3、 know Elastic Stack Can do 、 Nothing to do ?

4、 Positioning yourself , Development 、 framework 、 Operation and maintenance 、 Source code 、 Algorithm , Different positioning , Different learning paths .

but , If you want to go deep into , Source code is necessary .

5、 How to read 、 Watch more videos .

6、 More communication , Everyone's business scene 、 Knowledge system is limited .

Communication includes, but is not limited to : Exchange of experience , Project communication , Cognitive communication, etc .

4、 Powerful Alibaba cloud Elasticsearch

  • Simple —— Unlimited storage and computing power

  • To use —— Rich search function extension ( Face recognition ,dsl2sql)

  • cheap —— All kinds of scenario packages

  • Write efficiently —— Message queue cache , Improve the ability to write data .

  • Eyou—— Health monitoring , Detecting potential risks , Make development more convenient .

  • Separation of computing and storage —— Unwanted ES Multi copy mechanism , Rely on cloud storage

  • Cloud component —— participle 、SQL、 Vector Retrieval 、 flow control

5、 Summary

As Ruan Yiming said :“Elasticsearch At present, we are faced with : Control the cost 、 Improve accuracy 、 The challenge of maintaining high performance , In order to meet the challenge Elasticsearch Launched : Hot and cold architecture 、 Cross cluster replication 、 Snapshot search , And the meeting mentioned : Asynchronous search 、Off heap Other characteristics , Compared with other big data products ,Elasticsearch Catering to the trend of big data era .” therefore ,Elastic It's developing very fast , Version update and iteration are fast !

As Elastic Of iron powder , Every time I see a new feature 、 Technical points appear ( such as : Recent Wildcard Type、Date Streams etc. ), Will be excited .

At the end of the round table , Mr. Wang Gang asked each guest to make a summary :

Medcl teacher : choose Elasticsearch That's right !

Teacher Ruan Yiming :Elasticsearch, You deserve it !

Mr. Li Meng : Any data problem can be used ES To dismantle , If not , Another set, please !

This summary as the end of this article is also very close to .

6、 Egg ring

  • Video suggestion :2 Speed up , pretty 2 All over , That's interesting ( I'm very happy ).

  • Mr. Wang Gang and medcl The list of books in the teacher's study , Can you provide me with ( I only saw one 《 This is the search engine 》)? To the general Elastic Enthusiasts provide an introduction to this area .

  It may be the unofficial one with the most attention in China Elastic official account !

版权声明
本文为[Mingyi world]所创,转载请带上原文链接,感谢

Scroll to Top