Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage News Microsoft's Approach to Modern Data

Microsoft's Approach to Modern Data

This item in japanese

The PASS Summit 2017 is a conference for users of SQL Server and related Microsoft data technology.  At Day 1 of the conference, Microsoft's Rohan Kumar was on hand to present the opening keynote and take the opportunity to discuss the latest advances Microsoft has for SQL Server and Azure-based databases.

Kumar’s talk was intended to address the primary three themes data, AI, and Cloud.  He described the “Modern Estate” wherein data stores range from private, on-premise data stores to the public Cloud, and the hybrid area in-between.  Kumar sampled the conference attendees (by a show of hands), and it was clear that most firms are planning on a hybrid cloud approach.  

Recapping some statistics on the SQL Server 2017 adoption rate, Kumar noted that Microsoft saw 2M previews lead up to General Availability, with 500k trials since General Availability began just over a month ago at the Microsoft Ignite conference.

Several demos took place throughout the keynote to highlight some of the key aspects of SQL Server 2017 and related technology.  A forthcoming HP Gen10 DL580 offers persistent memory support, and the speed increase moving from SSD to in-memory queries was demonstrated to decrease to about 2 seconds from 15.  Highlighting another aspect of the SQL Server, the data base was running on SUSE Linux Enterprise Server.

Another benefit of SQL Server 2017 is its introduction of Automatic Tuning.  In this scenario, the database identifies performance degradations, in real time, on running queries and makes corrections to fix performance regressions without the involvement of a DBA.

Kumar observed that approximately 2 million live DBs on Azure at present.  To further increase adoption, Microsoft is introducing a new deployment option which offers Azure SQL Database Managed Instance – entire SQL instance is managed in Azure.

Azure SQL Database Features

  • Intelligent DBaaS
  • Privacy and Trust
  • Seamless and compatible
  • Competitive TCO

Looking to improve on the Azure Data Factory originally introduced at Ignite, Kumar said that one major areas for improvement identified by users was the inability to easily model data that didn’t involve a time series.   To that end, the improved Azure Data Factory was announced:

The New Azure Data Factory

  • Easily model diverse data integration scenarios
  • Create & manage data integration across on-premises, cloud & SaaS
  • Data movement as-a-service with 30 connectors
  • Easily move your SQL Server Integration Services (SSIS) workloads to cloud (Preview)

SQL Server Integration Services in Azure

  • Azure Data Factory now provides a managed environment for SSIS execution
  • Pick number of nodes and nodes size
  • Same SSIS runtime across Windows, Linux, and Azure Cloud

Scott Currie was on hand to demonstrate using Azure Data Factory with BIML (Business intelligence markup language) and noted that Azure Data Factory V2 has full support for BIML attributes.

Rate this Article