Elasticsearch is a NoSql database optimized for full text search.
It’s the leading Open Source search tool for optimized website search used on major websites like The New York Times, Shopify or Vimeo.
Elasticsearch is built for scalability, that made it an important piece for big data, opening a lot of use cases with log management or Business Information.
What is Elasticsearch ?
Elasticsearch is the main layer of the Elastic stack.
It’s an NoSql database, optimized for search, scalability and near real time data management.
Created in 2004 as an Open Source project based on Apache Lucene, Elasticsearch has grown over the years as a leading tool in the open source community.
It’s still leading by the Elastic team who offers paid featured, support, cloud hosting and training.
As a text search tool, with scalability and speed in the ADN of the project, the solution has been adopted by majors companies all over the world for several use cases :
- Salesforce : for logs management
- Netflix for search and log management
- Ebay for search and log management
- Vimeo for search and log management
- Github for search
- Bnp and Société Générale for search, security and fraud detection
- And also Orange, Axa, OUI SNCF, Facebook help, Uber, Tinder, The New York Times, and so many more.
Elastic nodes (server) are designed to work in clusters to ensure high availability and speed for all vital business use cases.
How Elasticsearch can help you ?
At first, Elasticsearch is… for search.
Nowadays, a basic search consists of a phrase typed within a smartphone in a simple search box, usually with some typos.
Elasticseach can return typo tolerant results based on the similarity of the search, and order by a score of similarity.
And it can do it really fast!
Elasticsearch can also do aggregations and search in only one request, returning contextual categories and the number of items in it, the quickest way.
Speed and relevancy of search is an essential point of the ROI of your site.
Clearly, for e-commerce websites, news sites, and any other websites with a lot of content, it’s vital!
Default Elasticsearch behavior does a pretty good job, but it can be enhanced to respond to each business specific needs.
But you should definetly consider adding advanced features to drastically improve the interest of an application for end users such as :
- ‘similar articles’ search,
- geosearch, searching in word or pdf content,
- the ability to tweak the request to return close results, even if the user search returns nothing
- searching in multiple source at one time
Scalability is in the core of Elasticsearch.
A cluster of Elasticsearch nodes (server) can be really big, allowing you to manage a huge amount of data.
And the engine is optimised to parallelise handling of requests to share the loads to all loads to ensure quick answers, even with several TB of data.
Business Intelligence (BI)
The ability for quick manipulation of a lot of information, makes Elasticsearch a perfect database for your BI needs.
Very complex aggregation requests can be used to build real time dashboards, enabling companies to multiply the power on all their business information.
Then a lot of visualization tools are compatible with elastic, and SQL-like endpoints make it possible to interface it with nearly everything.
To facilitate the graphic creation, Elastic also offers the Kibana tool to create impressive visualizations in a few clicks.
> See Spoon consulting’s Business Intelligence uses cases
Log management / security
Logs are by far one of the most important sources of data of any company.
They can say a lot about the behaviors of end users, network loads of security attacks.
The large realtime capacities of the Elasticsearch database make it a perfect tool to manage time based entries like logs, to analyse it and to show easily the abnormal behaviors.
Elasticseach can then send an alert to your admin and warn some other security tools to automatically block the threats.
With Logstash as lightweight ETL, Beats datashippers, Ingest nodes and pipelines, Kibana for visualisation and machine learning tools, Elastic gives a powerful Stack for all log management needs.
Spoon consulting is a certified partner of Elastic
As a certified partner of the Elastic company, Spoon Consulting offers a high level consulting for all kinds of companies, ranging from small and medium companies to large multinationals.
More than 90% of our clients use Multi-cloud today. Besides, many of them use more than one public cloud provider. In this context, more than ever what is the most important is : data quality, adoption of cloud and integrity with the SI.
Read more information on your personal use Elasticsearch use case on Spoon consulting’s posts