Azure Hdinsight will support Hadoop 2.4 performance improvement 100 times times

June 3, 2014, http://www.aliyun.com/zixun/aggregation/13357.html ">azure Hdinsight released an update message that Azure Hdinsight will support Hadoop 2.4, and improve the performance of the query data 100 times times. Today, we are announcing the beginning of the preview of the Apache HBase cluster (cluster) in the Hdinsight product.

HBase is a low latency NoSQL database, suitable for online transaction processing of large data (OLTP, online transactional 處理). We provide HBase clusters in azure, and the clustering data is stored directly in Azure BLOBs, which enables maximum access efficiency and flexibility in cost and performance considerations. This product allows our customers to have a large database to build highly interactive Web sites, or to store telemetry data returned from million sensors, and to analyze the data in Hadoop.

How to build a HBase cluster

Since the HBase on Azure Hdinsight is still in the preview phase, it is necessary to use PowerShell in operation.

PowerShell installed for Azure environment optimization

Follow the article steps to set the environment

The authentication data for Azure is in the variable using the following directives:

PS c:\> $creds = get-credential

Create HBase clusters (you can modify the data center location as needed and modify the Blob account data):

PS c:\> new-azurehdinsightcluster-name yourclustername-clustertype hbase-version 3.0-location "West US" Defaultstorageaccountname Yourstorageaccount.blob.core.windows.net-defaultstorageaccountkey " Yourstorageaccountkey "'-defaultstoragecontainername hbasecontainername-credential $creds-ClusterSizeInNodes 4

Manipulating data in a HBase cluster

Application developers can access HBase data through REST APIs, HBase shells, or different map/reduce tools such as Hive and Pig, and the HBase Shell provides an interactive console (console) that allows you to manage HBase clustering, creating or deleting data tables (table), and manipulating data.

To use the HBase shell, you must first open the established HBase cluster Remote Desktop Online (RDP) to connect to it.

At the end of the cluster setup, you can set the revisit page on the Azure management interface, press the ENABLE REMOTE button at the bottom to open the RDP, and use the CONNECT button to go online.

After linking into the cluster, press the shortcut to the Hadoop command prompt on your desktop, and then enter the following instructions to open the HBase shell:

CD%hbase_home%\bin

HBase Shell

The following instructions create a sample data table, add a column of data, and list all the data in the datasheet:

Create ' sampletable ', ' cf1′

Put ' sampletable ', ' row1′, ' cf1:col1′, ' value1′

Scan ' sampletable '