Using Azure Data Lake Storage

Data Lake Storage is the data storage component of the Azure Data Lake service. Data Lake Storage (middle) is optimized to receive unstructured data at high speed in its native format. It can accept files ranging to petabyte size, and the data storage component can scale without limit as needed. This flexibility allows the service to work as a general data storage solution, regardless of data type.

Data Lake Store has integrations and connectors to a wide array of Azure services and other tools for ingestion and query. It can accept data from a variety of data providers (bottom), including several Azure data services and Hadoop Distributed File System (HDFS)-compliant tools. Customers can populate Data Lake Storage with Azure data capture services, such as Azure Event Hubs and Azure Stream Analytics, or they can use on-premises and online HDFS tools to populate Data Lake Storage with existing processes and applications.

The stored data can be queried and analyzed (top) using Azure Data Lake Analytics (the query component of the service), HDFS-compliant tools, and other Azure-based business intelligence (BI) and analytic services, including Power BI, Azure HDInsight, and Azure Machine Learning. This capability could make the potential of machine learning available to more organizations by leveraging Azure’s cloud-computing model and simplifying the process of working with machine learning algorithms and methodology. (Data Lake Analytics is discussed in “Using Azure Data Lake Analytics“.)