Columnar data stores

Designed for analytics, columnar data stores organize data in columns rather than rows. This makes aggregations and analytics blazing fast, as reading becomes more efficient for analyti-cal queries.

For example, imagine a scenario where you’re analyzing sales trends based on product colors across millions of entries. Rather than scanning rows, columnar stores read the color column efficiently, as shown in Figure 4-6.

Skill 4.1 Describe common elements of large-scale analytics     CHAPTER 4       111

FIGURE 4-6  Row-based versus columnar storage

Time-series data stores

When you’re dealing with time-stamped data, such as stock prices or sensor readings, time-series databases come into play. They’re optimized for chronological insights.

For example, picture a weather station collecting temperature data every minute. Over months, this becomes a massive dataset, best stored and analyzed via a time-series database. Figure 4-7 shows temperature data collected over a week.

FIGURE 4-7  Time-series data store

When you dive into Azure’s analytical ecosystem, you’ll encounter two standout data stores tailor-made for your analytics needs: Azure Synapse Analytics and Azure Data Lake Storage Gen2.

With Azure Synapse Analytics, previously known as Azure SQL Data Warehouse, you get a compilation of big data warehousing capabilities. This platform allows you to swiftly scale resources up or down, paying only for what you use. An integrated workspace in Synapse Ana-lytics offers you a singular experience, blending big data and data warehousing. Whether your workload demands on-demand query processing or provisioned resources, Synapse Analytics

112 CHAPTER 4     Describe an analytics workload on Azure

accommodates both. What’s more, Synapse Studio lets you explore your data on the fly, without the hassle of infrastructure setups. Azure Data Lake Storage Gen2 is a vast, scalable, and secure data lake optimized for big data analytics. The unique aspect of this store is its hierarchical namespace, which combines the prowess of Azure Blob Storage with a structured file system, thereby boosting analytics performance. Security is paramount here. Integration with Azure Active Directory, combined with POSIX-compliant access controls, ensures your data’s sanctity. To top it off, tools like Azure Data Factory come to your aid for seamless data movement, making it easier for you to inte-grate with other Azure data services such as Azure Databricks or Azure HDInsight.