Cloud usage is on the up, for example, with 44% of organizations now using SQL Server on VMs in Azure, Amazon and the Google Cloud at least some of the time, 38% using Azure SQL Database, and 23% using Azure Managed Instances.
Alongside this development, SQL Server 2008 is coming to the end of its life and the cessation of bug fixes, security updates and mainstream support is prompting organizations to upgrade to a newer version of on-premises SQL Server.
As Jamie Wallis, Redgate Product Marketing Manager, comments: “In a way, this is a perfect storm for SQL Server professionals. The organizations they work for want the advantages the cloud brings, particularly for greenfield projects, and they also want to avoid the risk of using an unsupported platform. The task is to migrate to and manage what are now becoming hybrid server estates, yet still be able to quickly and easily monitor them for any performance issues that come up.”
This is further complicated by the growth of SQL Server estates. Those with fewer than 10 servers now make up 33% of the market, compared to 37% last year, for example, and those with 10-49 servers have also fallen by an average of 4%. In contrast, there has been in increase of around 4% in those with estates of 50 servers and more – estates which in some cases now exceed 1,000 servers.
The rise in the frequency of database deployments also brings its own problems. 73% of organizations are now deploying database changes multiple times a month, and 47% are deploying changes multiple times per week. 41% also expect this to continue to increase over the next 12 months.
This may well be leading to the top cause of database issues highlighted in the report: human error. Quite simply, as the pace of database development and the frequency of deployments goes up, so faults creep into the development process which need to be identified and corrected.
It’s perhaps no surprise then that 43% of respondents to the survey now use a third-party monitoring tool, a figure which rises to 60% in organizations with 50 SQL Servers or more. As well as providing long-term tracking of trends like growth in database size, they can identify issues with queries where human error has caused problems, fix them, and prevent them happening again.