Getting Started with OLAP on Azure

Author
Coco Li
Product Marketing Manager
Jul. 20, 2022
   

This article will help you to understand the deployment of OLAP on top of Azure, including data sources, features, benefits, and prerequisites. The combination of features creates a number of different cost and usage options and, depending on your needs, you can create the right OLAP on Azure solution for your organization.

 

Why companies shift to OLAP on Azure

 

Online analytical processing (OLAP) is a system for performing multi-dimensional analysis at high speeds on large volumes of data, typically sourced from a data warehouse, data mart, or some other centralized data store. OLAP is ideal for data mining, business intelligence, and complex analytical calculations, as well as business reporting functions like financial analysis, budgeting, and sales forecasting.

 

Currently firms have two options to implement a data warehouse: one is on-premises relational database management system (RDBMS)-based, and the other is cloud-native.

 
RDBMS-based data warehouse
 

An RDBMS-based data warehouse is a mature product and proven methodology that can meet an organization’s data management and analysis needs when using relatively small data volumes. However, it relies heavily on proprietary on-premises servers, which come at a high cost and make it difficult to scale. And, once you decide on a proprietary product from a specific vendor, options for future migration are limited and costs can be very high.

 
Cloud-native data warehouse
 

In Azure, data stored in online transaction processing (OLTP) systems (e.g. Azure SQL Database) is replicated to OLAP systems (e.g. Azure Analysis Services). Data exploration and visualization tools (e.g. Power BI, Excel, and third-party options) connect to the Analysis Services server, allowing users to understand the modeled data in an interactive and visually rich way. Data flow from OLTP to OLAP is typically arranged using SQL Server Integration Services (SSIS) or the Azure Data Factory.

 

the following data stores all meet core OLAP requirements:

 
  • SQL Server with column store indexes
  • Azure Analysis Services
  • SQL Server Analysis Services (SSAS)
 

SQL Server Analysis Services (SSAS) provide OLAP and data mining capabilities for business intelligence applications. SSAS can be installed on a local server or on a host within a virtual machine in Azure.

 

Azure Analysis Services is a fully managed service that provides the same key functionality as SSAS. Azure Analysis Services supports connections to a variety of data sources in the cloud, and local data sources in the organization.

 

A cloud-native data warehouse in Azure achieves the balance between operation efficiency and cost. With one-click deployment, managed operation, and on-demand scaling, it greatly lowers the total cost of ownership (TCO). But cloud-native data warehouses still suffer from a prolonged data development cycle. Here is a typical development workflow:

 
  • Data analysts initiate an analysis request and then collaborate with data engineers for a solution.
  • Data engineers put lots of effort into development and performance tuning.
 

Despite the lengthy development process, the solution cannot be directly reused, and labor costs increase with business growth. Fortunately, organizations that choose to address their analytics needs by using Kyligence’s OLAP on Azure can avoid these shortcomings and achieve more consistent and efficient results at a lower TCO.

 

Kyligence is changing the game by offering a way for operations engineers, data engineers, and data analysts to increase their productivity and liberate themselves from legacy data warehouse solutions. Harnessing cloud-native computing and storage resources, Kyligence enables fast, elastic, and cost-effective analysis innovation with any data lake and at any scale. Kyligence's AI-augmented engine detects patterns from the most frequently asked business queries, builds governed data marts automatically, and ensures metrics consistency on the data lake to optimize data pipelines and avoid an excessive number of tables.

 

Features of OLAP on Azure 

 

OLAP on Azure-based applications have a variety of enterprise-class features. Combining the above best practices, Kyligence’s OLAP on Azure solution delivers four compelling factors to consider for implementing OLAP on Azure:

 

Auto scaling

By separating computing and storage, OLAP on Azure offers a low-cost auto-scaling solution to satisfy ever-growing data analytics demands.

 

High performance query response

OLAP on Azure provides high-performance, precomputed result sets that deliver sub-second query response times against very large datasets.

 

High compatibility

OLAP on Azure is compatible with ANSI SQL specifications, and provides a standard SQL query interface.

 

Dimensional Modeling

Although Kimball's dimensional modeling has a high learning threshold, cloud-based solutions can simplify dimensional modeling through the power of AI.

 

Front end support

With  REST API, JDBC Driver, and ODBC Driver, OLAP on Azure can easily integrate with third-party applications, such as Tableau, Power BI, and other data analysis applications.

 

Security and reliability

With private network and full SSL encryption links, OLAP on Azure maximizes data security and reliability together with security group strategy on the cloud.

 

Use cases for OLAP on Azure

 

There are two typical use cases to consider when evaluating OLAP on Azure.

 
Analytics on Data lake
 

In the context of cloud migration, enterprises often choose object storage of high reliability and low cost as a single data storage pool to store their structured and unstructured data.

 

Kyligence Cloud supports seamless integration with object storage from different cloud vendors to provide a single source of truth. Computing resources can be independently applied to applications on data lakes. What’s more, Kyligence Cloud also supports on-demand start-stop to help lower TCO.

 
Seamless Azure Analysis Services upgrade
 

Azure Analysis Services performs well with small-scale data, but may become slow when using massive data volumes, or in high-concurrency scenarios, because of its scale-up architecture.

 

Kyligence Cloud fully utilizes separating compute-and-storage design. When facing high-concurrency analytics requests, Kyligence Cloud performs auto scaling of computing and storage resources to achieve sub-second response times.  

 

Benefits of OLAP on Azure

 
  • One-click deployment: Simplify and automate the deployment process on Azure cloud.
  • Unified semantic layer: Unified semantic layer maps complex data into familiar business terms. Users can directly consume the business definitions and reuse the semantics in different downstream applications.
  • Enterprise-grade security: Kyligence Cloud provides enterprise directory integration, role-based authorization, and cell-level access control.
  • Intelligent storage optimization: Intelligently identify low-efficiency storage, and optimize according to ROI.
  • Lower cloud costs: Kyligence's elastic resource allocation saves resources and lowers costs. Data storage is separate from computing resources, clusters can be started on-demand, and compute resources are only provisioned when necessary. In addition, when load sizes decrease, the cluster will automatically decrease.
 

How to get started with OLAP on Azure

 

If you want to increase the efficiency of your organization’s analytics with Enterprise OLAP on Azure, why not experience Kyligence's OLAP on Azure solution for yourself? You can try us for free with a trial Test Drive; or place a self-service order through the Azure marketplace

   

For more information about OLAP on Azure, read about how to Accelerate BI on Big Data with Kyligence, learn more about Kyligence for Azure, or contact us directly with questions specific to your organization’s needs.