Introduction: This is a continuation of a series of articles on Azure services starting with the SignalR service mentioned here. The Azure analysis service is next in the portfolio of Azure cloud services we started to review. This is a fully managed platform-as-a-service that provides enterprise-grade data models in the cloud. It uses advanced mashup and modeling features to combine data from multiple data sources, define metrics, and secure the data in a single trusted tabular semantic data model. The data model provides an easier and faster way for users to perform ad-hoc data analysis using tools like PowerBI and Excel.

Azure analysis service is different from other services in that, it provides a one-stop-shop for all tabular data models across hybrid data sources and even at higher compatibility levels than others. Tabular models are key concepts of relational modeling and are articulated by definitions in tabular model scripting language as well as tabular object models. With the help of tables we can define partitions views row-level security bidirectional relationships and translations. Since the data can be accumulated across cloud and on-premises data sources, it provides a consistent and enterprise-grade semantic data model which can be used with power tools like power BI and excel.

Performance of analysis services is well known by virtue of its support for partitioning enables loads to be increased incrementally, queries to be run parallelly and, memory to be consumed in an efficient manner. There are also advanced data modeling features like calculated tables and all DAX functions. Regardless of the size of the data that is transferred to the server, the analysis service supports in-memory models that can be refreshed with cache data or data directly from the sources. It offers support for recognizing Azure logged-in users and service principals so that querying and analysis can be run under their security context which enables all the auditing benefits.  Background operations and unattended refresh can be automated, it offers a variety of languages to integrate with its SDK, REST APIs, and PowerShell cmdLets. In comparison to a standalone SQL Server analysis service, this cloud service offers unprecedented virtualization of both relational databases as well as warehouses, which makes querying and analysis quite easy.

The service is hosted on a single server, and it can scale to several servers with simple deployment techniques that are common to many Azure services where the infrastructure can be provisioned declaratively using ARM resource templates. The Azure analysis service is supported in regions throughout the world with varying types of SKUs and pricing options. With the help of replicas, this Azure service can scale-out queries in a distributed manner and still maintain relatively low latency in query responses. It does this by creating a query pool with up to seven additional query replicas and these replicas would be assigned to the same region, it is quite possible to dramatically increase performance with premium and higher sizes of the deployments.

The analysis service stands out by itself among its peers of analysis services and pipelines produced from a combination of heterogeneous products by being a native cloud service. It conforms to the strict and stringent demands from the cloud computing provider and thus meets industry standards and government compliance requirements in terms of privacy, security, and data protection.