Create and Configure Azure HDInsight (original) (raw)

Last Updated : 23 Jul, 2025

In our chapter about the amazing Poly Base thingy, we presented this super cool SQL Server 2024 feature to query CSV files stored in Azure Storage accounts. We mentioned that in PolyBase, hey, you can query data in Hadoop (HDInsight) using SQL Server. HDInsight is like, totally a very popular system in Azure that eventually you will, like, need to interact with if you use SQL Server. That is why we will, like, give an explanation for all the newbies out there about it, you know?

What is Hadoop?

It's an extremely scalable Distributed File System (HDFS) used for handling big data. There are multiple scenarios when a traditional database such as SQL Server or Oracle is not the optimal way to store data. For instance, to store YouTube or Facebook info, it would be very expensive to store all the images and videos in a traditional database. That's why Hadoop was invented. Hadoop can handle Petabytes of info easily using several distributed computers. With Hadoop, you can easily manage SQL and NoSQL Data and it's easy to distribute the info to several servers.

What is HDInsight?

• In popular generation for huge statistics analytics is Apache Hadoop. Large volumes of historical or flowing records may stored, processed, and analyzed with the useful resource of Hadoop. Additionally it has the potential to be scaled up as needed. By a resenting a one-forestall keep, Azure HDInsight makes it less complicated for us to method huge data the usage of open-source frameworks like Hadoop.
• For Using the open-source frameworks for big statistics analytics made viable through Microsoft's Azure HDInsight provider. Azure HDInsight permits using frameworks like Hadoop, Apache Spark, Apache Hive, LLAP, Apache Kafka, Apache Storm, R etc. For processing massive quantities of facts. These equipment can be used for information warehousing, device getting to know, and extraction, transformation, and loading "ETL".

Understanding Of Primary Terminologies

• **Azure HDInsight: An azure service that provide managed clusters for big data processing and analytics.
• **Hadoop: A Distributed open source processing framework helps in processing large datasets across the clusters using simple programming models.
• **Apache Spark: It is an open source distributed computer system that used for providing an interface programming entire all clusters with implict data parallelism and fault tolerance.
• **Apache Hive: It is built on top of Hadoop provides data summarization, query and analysis acting as a data ware house infrastructure.
• **Cluster: A group of inter connection devices that performing the work as single unit.
• **Blob Storage: A type of Azure storage service that is used of storing a large amount of unstructured data such as text or binary data.
• **Data Lake Storage: A scalable and secure storage service from Azure provider that is used for big data analytic workloads.

Configuring Azure HDInsight : A Step-By-Step Guide

**Step 1: We will learn how to create an Hadoop clusters, upload a CSV file and query the file using HIVE (a query language in Hadoop)

• You can have various cluster types like Hadoop, H base, and Storm. For this example, let's select Hadoop! You can use Linux or Windows, like the Hadoop OS; you can also choose the Hadoop versions.
• In Cluster Tier, you can pick the Standard and Premium tiers. The Premium Tier is pricier and encompasses AD Integration and secure Hadoop, (Ranger)!

Step 2: Cluster Configuration

• You can create many various configurations. Hadoop are the traditional cluster.

• H base used is for Columnar NoSQL data, Storm used for stream Analytics for real-time processing Spark is for In-memory interactive queries and micro batch, stream processing interactive Hive is used for queries In-memory and caching R Server is mainly used for machine learning tasks!

Cluster Type

**Step 3: In a credential section, you will need login access and administer the cluster and another account to use SSH.

• SSH is a secure shell to administer remote Servers from a local file using the command line.

**Step 4: HDInsight is stored in an Azure Storage Account; it's then stored in a container. A container is kind of like a folder to store information in Azure. You can also specify the location to store. Usually, the location should be, close to, your local, location!

• In a Pricing is very important. More cores and RAM will increase the price!!! You have working nodes, which contains the data and information and the head nodes; which are used to host the services.

**Step 5: Then Press View all to see all the different options:

**Step 6: You need create resource group or create new one. There groups are used to a group resources to the make the administration easier and the press create:

Resource group

**Step 7: The Login with the credentials a created and press Log In

**Step 8: In a Dashboard shows hardware of information like disk usage, node time, number of live nodes, memory and network usage:

**Step 9: Go Query Tab and click Default. The table created by default hive sample table:

**Step 10: Can query the customers csv file using the following query:

**Step 11: In results you see the values of the csv File like if were table:

**Step 11: That the MASE is installed, connect to the Azure Storage Account and blob container created in Step 4 and go to the hive folder:

Conclusion

In a Azure HDInsight is a robust cloud service that empowers organizations to unlock the potential of big data by offering a fully managed environment for Apache Hadoop and Spark clusters. By understanding its features, configuration process, supported cluster types, and data storage options, users can harness the power of Azure HDInsight to drive meaningful insights and innovation in their data analytics endeavors.