During the past few decades, data warehouses have become the core of an enterprise's information and decision support...
systems. These central repositories contain all relevant data (internal or external) for enterprise information and decision making. However, in today's world of interconnected devices, such as smartphones, televisions, watches, laptops, tablets, desktops and gaming devices, as well as wide access to data obtained from various sources such as Twitter, Facebook, LinkedIn, flat files, blogs, websites, system logs and sensors, data growth is among one of three main challenges companies face today. As a result of exponential data growth within organizations, traditional data warehouses have reached a critical point -- one that will require major investments in hardware, tuning, support and maintenance.
Moreover, in recent years, organizations are using Apache Hadoop to process big data from sources such as blogs, sensors, social media, system logs and other devices. However, traditional data warehouses do not allow end users to query structured data along with unstructured data. That means end users cannot collect and analyze data, regardless of size and type. In addition, traditional data warehouses are not optimized for low-latency large volume data load and high-throughput complex analytical workloads -- a requirement for meeting the demands of big data.
Analytics Platform System introduced as a modern data warehouse
To meet enterprise demands and help organizations transition to a modern data warehouse optimized for low-latency large volume data load and high-throughput complex analytical workloads, in April 2014 Microsoft introduced the Analytics Platform System (APS), also known as the Parallel Data Warehouse (PDW). APS is a high-performance and scalable parallel processing appliance built for modern data warehousing needs. The certified hardware platform integrates SQL Server PDW software (a massively parallel processing version of SQL Server designed to run within APS) and an optional HDInsight Hadoop platform (a Microsoft offering of Hadoop for Windows based on the Hortonwoks Data Platform) together in the same appliance. Big data capabilities in APS, with the included PolyBase, let you perform standard SQL queries to access and join Hadoop data with relational data, without having to preload data into the data warehouse. This seamless integration between traditional data warehouses and big data deployments makes APS a leading enterprise-ready big data platform.
APS also supports new scenarios for using Power BI modeling, visualization and collaboration tools with on-premises data sets. For example, the native Microsoft BI integration allows end users to analyze relational and non-relational data with familiar tools such as Microsoft Excel.
New features in Analytics Platform System
APS can handle the extremes of the largest mission-critical requirements because it is a massively parallel processing appliance that parallelizes and distributes computing for high-query concurrency and complexity. SQL Server PDW, which runs inside the APS appliance, improves data load and query response times up to and beyond 50 times over legacy data warehousing systems through in-memory and updateable columnstore indexes, allowing queries from end users to complete in minutes rather than hours, or in seconds rather than minutes.
APS has resilient, scalable and high-performance storage features built into software, which lowers hardware costs -- and it has built-in hardware redundancies for fault tolerance. The appliance also helps organizations reduce data center and management costs because it combines a relational data warehouse with Hadoop. APS provides data compression up to 15 times with in-memory updateable columnstore indexes, saving up to 70% of storage requirements.
APS is a rack-based system, so rather than over-acquiring capacity, you can begin with a quarter rack, which lets you correctly size the appliance and then scale out at a later date with the same tools that you normally use when scaling out traditional SQL Server systems. Microsoft co-engineered APS with Dell, HP and Quanta, and Microsoft is your single point of contact for hardware and software support. You will find that, as a data warehouse appliance, APS offers the lowest price per terabyte for user-available storage (compressed).
Find out more about the Application Platform System and the vendors selling the appliance
Learn more about the different versions of the Parallel Data Warehouse Appliance