Netherlands: Software

Introductie van Micorosoft SQL Server 2016

Issue link: http://hub-nl.insight.com/i/692679

Contents of this Issue

Navigation

Page 52 of 212

41 C H A P T E R 3 | Higher availability Figure 3-5: Log transport operations for AlwaysOn Availability Groups. Bottlenecks can occur in this process during the log-capture step on the primary replica and the redo step on the secondary replica. In previous versions of SQL Server, both steps were single-threaded. Consequently, bottlenecks might occur during large index rebuilds on availability groups with high- speed storage and on local networks, because these single-threaded steps had trouble keeping up with the stream of log records. However, in SQL Server 2016 these steps can use multiple threads that run in parallel, resulting in significant performance improvements. Furthermore, the compression functions in the log-capture step have been replaced by a newer Windows compression function that delivers up to five times better performance. During testing with high-throughput storage devices, speeds up to 500 MB/s have been observed. Considering that this throughput is a compressed stream, the redo step is receiving 1 GB/s, which should support the busiest applications on the fastest storage. Microsoft Azure high-availability/disaster-recovery licensing changes Hybrid disaster-recovery scenarios are becoming increasingly popular. If you choose to implement hybrid disaster recovery, be sure to maintain symmetry between on-premises and cloud solutions. The license mobility benefit included with software assurance (SA) allows you to use a secondary copy of SQL Server in any high-availability or disaster-recovery scenario without purchasing another license for it. In the past, you could not use this benefit with the SQL Server images on Azure Virtual Machines. Now you can deploy a new SQL Server image on an Azure Virtual Machine without incurring charges as long as the secondary replica is not active. This means you can automate the scale-out of your high-availability/disaster-recovery solutions with a minimum of effort and cost. Windows Server 2016 Technical Preview high- availability enhancements Nearly every version of Windows Server since Windows Server 2008 R2 has had major enhancements to the operating system's failover clustering stack as a result of development investments in related technologies. First, Hyper-V, the virtualization platform in the operating system, uses the clustering stack for its high-availability and disaster-recovery scenarios. Microsoft Azure also uses this same functionality. Because SQL Server has failover clustering at the center of its high-availability/disaster- recovery technologies, it also takes advantage of the clustering features in the operating system. Sometimes these features are visible from the database tier, allowing you to make configuration

Articles in this issue

Archives of this issue

view archives of Netherlands: Software - Introductie van Micorosoft SQL Server 2016