Microsoft has released Community Technology Preview (CTP) 1.3 of the next version of SQL Server for both Windows and Linux, codenamed SQL Server vNext.
The newest preview release follows last month's CTP 1.2, which included SUSE Linux support. This time around, the software giant added the Always On Availability Groups feature to the Linux edition, enabling customers to test the database software's hardiness while running critical workloads.
"Always On Availability Groups is SQL Server's flagship solution for HADR [high availability and disaster recovery]," explained Mihaela Blendea, senior program manager of SQL Server at Microsoft, in a blog post. First appearing in SQL Server 2012, the feature "provides High Availability for groups of databases on top of direct attached storage, supporting multiple active secondary replicas for integrated HA/DR, automatic failure detection, fast transparent failover, and read load balancing. This broad set of capabilities is enabling customers to meet the strictest availability SLA [service-level agreement] requirements for their mission-critical workloads," she added.
Always On Availability Groups works on all supported Linux distributions, namely Red Hat Enterprise Linux, Ubuntu and the aforementioned SUSE Linux Enterprise Server. It enables multi-database failover, fast failure detection and failover and database-level health monitoring and failover triggering, among other capabilities designed to help keep critical applications up and running in the event of an outage or other mishap that strikes a customer's data center.
An overview of the feature and its use cases is available in Blendea's blog post.
Also new in SQL Server vNext CTP 1.3 is full text search in all supported Linux distributions, resumable online index rebuilds and support for Temporal Tables Retention Policies that eases the management of historical data stored in temporal tables. Indirect Checkpoint, a recommended configuration used in large databases, performs faster courtesy of new tweaks.
Another performance-enhancing feature is Encoding Hints. In a separate post, Christian Wade, senior program manager of Analysis Services at Microsoft, explained that the feature can be "used to optimize processing (data refresh) of large in-memory tabular models."
Typically, when running Analysis Services on SQL Server, the feature takes sample values from a numerical column to choose between a data encoding method (value encoding or the space-saving hash encoding method). After partially processing a column and gathering enough information about how data is distributed therein, Analysis Services can switch gears, requiring a time-consuming restart of the encoding process, said Wade.
Encoding Hints may help speed things up by allowing "the modeler to specify a preference for the encoding method given prior knowledge from data profiling and/or in response to re-encoding trace events," added Wade. "Since aggregation over hash-encoded columns is slower than over value-encoded columns, value encoding may be specified as a hint for such columns."
Encoding Hints is not a setting, he cautioned, since there are no guarantees the software will apply the "hints." A complete list of new functionality in CTP 1.3 is available in this online support document.