Microsoft releases the latest update to Analytics Platform System (APS)

Microsoft is pleased to announce the release of Analytics Platform System (APS) Appliance Update (AU) 5. APS is Microsoft’s scale-out Massively Parallel Processing fully integrated system for data warehouse specific workloads. Powered by industry-leading SQL Server as its core engine, APS leverages PolyBase, the tool that allows for integrated querying between relational and non-relational data technology, for Hadoop/Big Data integration.

The AU5 release offers customers greater Transact-SQL (T-SQL) compatibility to aid in migrations from SQL Server and other platforms as well as improved connectivity and integration with Hadoop. The AU5 release also includes support for faster data loading scenarios through both first party Microsoft and 3rd party tools. These features continue to provide greater alignment with SQL Server and bring significant value to our customers.

APS hardware partners Dell, HPE and Quanta Cloud Technology will ship APS with AU5 starting this month. Specific shipping dates will vary depending on the hardware partner’s factory process.

This update delivers several new capabilities and features, including:

PolyBase/Hadoop enhancements

  • Support for Hortonworks (HDP 2.3) and Cloudera (CHD5.5)
  • String predicate pushdown to Hadoop for improved performance
  • Support for National/Government cloud storage and public Azure data sets
  • Apache Parquet file format support

Data loading

  • BCP support for additional data loading scenarios
    • Supports the bcp.exe command line interface for simple data import/export scenarios
    • .NET SqlBulkCopy class support for custom application integration
  • Support for SQL Server Native Client OLE DB and the Bulk Copy API unlocking access by many 3rd party ETL, reporting and analytic tools

T-SQL compatibility improvements to reduce migration friction from SQL SMP

  • sp_prepexec: A common dynamic query preprocessor model that allows customers to simplify migrations
  • SET NOCOUNT and SET PARSEONLY set statements used across a variety of customer’s scenarios
  • IS_MEMBER() and IS_ROLEMEMBER() in support of Windows Authentication
  • CREATE TABLE as HEAP option allows customers to explicitly define heap, in addition to clustered index, or clustered columnstore index tables which aligns DDL with the SQL

Data Warehouse service

Based on early testing and feedback, we are seeing a performance improvement of up to 30% for short running queries.

In addition to the above, we are also offering an early preview of Adaptive Query Processing which can automatically re-optimize query execution mid-flight. Please note that Adaptive Query Processing is currently in beta. Any customers wishing to participate in the beta should contact their support representative.

Please visit the Analytics Platform System portal to learn more.