Expanding the data footprint of SQL Server 2016 with PolyBase 

3 min read

This post was authored by Casey Karst, Program Manager, SQL Server. Changing data landscape A lot has changed in the world of data over the last 10 years. The rise of connected devices, unstructured event data and ever-decreasing hardware prices has caused a Big Data boom. Solutions built on commodity hardware, such as Hadoop and Read more

2 Comments

Technical Overview: SQL Server 2016 Release Candidate 0 

5 min read

The SQL Server engineering team is pleased to announce the availability of SQL Server 2016 Release Candidate 0. This is an important milestone in the release of SQL Server 2016, as it marks feature completion for most dimensions of the product and means a very rich set of capabilities are now available. These include: real-time Read more

3 Comments

PolyBase in APS – Yet another SQL over Hadoop solution? 

7 min read

This blog post will highlight PolyBase’s truly unique approach focusing on: Query capabilities across various heterogeneous data sources – on-premises and cloud (Microsoft Azure) bringing Microsoft data services together forming one complete data platform solution Total freedom for users with no lockdown, agnostic to the actual Hadoop distribution and/or underlying operating system Faster insights from Read more

2 Comments

Change the Game with APS and PolyBase 

5 min read

Guest blog post by: James Rowland-Jones (JRJ), SQL Server MVP, PASS Board Director, SQLBits organiser and owner of The Big Bang Data Company (@BigBangDataCo). James specializes in Microsoft Analytics Platform System and delivers scale-out solutions that are both simple and elegant in their design. He is passionate about the community, sitting on both the PASS Read more

6 Comments

Microsoft’s Database Research Lab Conceives PolyBase to Easily Join Relational Data With Data From Hadoop 

1 min read

Complementing Microsoft’s overall Big Data strategy, PolyBase is a breakthrough new technology on the data processing engine in SQL Server 2012 Parallel Data Warehouse Appliance designed as the simplest way to combine non-relational data and traditional relational data in your analysis. While customers would normally burden IT to pre-populate the warehouse with Hadoop data or Read more

1 Comment