deep api integration makes getting value from your big data easy
深度api集成使你大数据訪问更加easy
Elasticsearch is quickly becoming the de facto search and analytics solution that organizations are using to provide real-time insights into their Hadoop data. Elasticsearch for Hadoop—affectionately known as es-hadoop—is a two-way connector that lets you index data into Elasticsearch and query it in real time. With a native API implementation, fast indexing, and a rich query language, es-hadoop is optimized for performance and efficiency, making it an elegant solution for your big data projects. With support for a wide range of libraries, Elasticsearch helps you to make better use of your data across the entire Hadoop ecosystem.
data can seamlessly move between Elasticsearch and Hadoop
- Index directly into Elasticsearch from Hadoop 直接对hadoop上的数据建立索引
The native integration allows you to efficiently push data into Elasticsearch using the existing Hadoop tools you know and love ,原生态的集成同意你通过你喜欢的hadoop工具将数据推送到ElasticSearch中
- Query Elasticsearch from Hadoop从hadoop查询Elasticsearch
The rich query API of Elasticsearch allows you to ask complex questions and use the real-time results in Hadoop.Elasticsearch丰富的查询api支持你迅速取得对hadoop的复杂查询结果。 - Use HDFS as a long-term archive for Elasticsearch使用HDFS对Elasticsearch索引长期存档
es-hadoop allows Elasticsearch to push backup data to HDFS using the built-in snapshot and restore capability.es-hadoop插件同意es推送备份数据到HDFS通过使用快照的方式和恢复这些数据到es
how people are using Elasticsearch and Hadoop
- Klout Queries Over 400M Users’ Data To Build Marketing Campaigns
Using HDFS to store user data and index it into Elasticsearch, Klout builds real-time targeted marketing campaigns that are generated in seconds rather than minutes. - MutualMind Replaces 15-Minute Batch Process with Real-Time Analysis
With customers like AT&T, Kraft, Nestle, and Starbucks interested in keeping a pulse on their brands, MutualMind uses Elasticsearch to get quick insight and Hadoop for batch-based statistical analysis. - International Financial Services Firm Quickly Analyzes Access Logs
Instead of waiting hours to run MapReduce jobs to analyze access logs, a global financial institution gets value from its data with Elasticsearch in minutes—and even increased the quantity of log data it processed from one hour to a full week.
works with any flavor of Hadoop distribution
![7698cfbc-4a89-11e4-89aa-05066fff2b59](http://www.elasticsearch.com/wp-content/uploads/2013/10/7698cfbc-4a89-11e4-89aa-05066fff2b59.png)
take a look under the hood
Elasticsearch works with the visualization tool Kibana to help you explore your big data with in real time. With beautifully designed graphs, charts, and maps, Kibana transforms your data into real-time, customizable dashboards that let you visualize the value of your data.
leave the real-time analytics to us
Gone are the days of waiting hours or more for a batch process to run in order to get insight into your Hadoop data. Elasticsearch provides responses in milliseconds, which can significantly reduce a Hadoop job’s execution time and the cost associated with
it, especially on “rented resources” such as Amazon EMR or EC2.
ask more sophisticated questions
Elasticsearch provides a robust query DSL that lets users to ask sophisticated questions that result in more complete answers, faster.
prepared for when things go awry
Elasticsearch is designed to tolerate hardware failures. Es-hadoop continues communicating with the cluster, even when failures occur.
added efficiency with our native integration
Elasticsearch is natively integrated with Hadoop so there is no gap for the user to bridge. We provide a dedicated Input and Output format for vanilla MapReduce, taps for reading and writing data in Cascading, storages for Pig and Hive, a native Spark Resilient
Distributed Dataset (RDD) for both Java and Scala, and support for Storm’s bolt and spout abstractions so you can access Elasticsearch just as if the data were in HDFS.
enhance your workflow to get the best of both worlds
Get maximum flexibility with the es-hadoop connector by leveraging everything that Hadoop has to offer (via MapReduce, Hive, Pig, Cascading, Spark, and Storm) and combining it with a real-time search and analytics capability of Elasticsearch.
need to grow? just add more nodes.
Elasticsearch can be scaled in the same way as your Hadoop cluster – add more Elasticsearch nodes and the data will be automatically re-balanced.
原文网址:http://www.elasticsearch.com/products/hadoop/