Shulou(Shulou.com)06/01 Report--

Compared with locally deployed databases, cloud databases can provide users with more flexibility and scalability, and the price is also lower. But cloud systems often have problems, so we must always pay attention to high availability and disaster recovery.

David Bermingham, a senior technology promoter from SIOS Technologies, detailed a set of SQL Server high availability best practices for cloud users at a webinar hosted by SQL Server Cluster Management Software, while comparing the high availability features provided by Microsoft Azure, AWS, and Google clouds.

Bermingham says reliable high availability policies can help create efficient workflows, uptime, and effective disaster recovery procedures for cloud-based SQL Server systems. If the alternative is not ideal, there may be database outages, data loss, and other headaches for SQL Server database administrators (DBA).

Here are some Bermingham recommendations for SQL Server high availability best practices.

Read the details before using the cloud platform

Azure's service level agreement (SLA) promises only 22 minutes of database downtime per month. However, according to Bermingham, this occurs only if two or more instances of SQL Server are grouped into an availability set consisting of multiple virtual machines (VMs). If you set up only one instance yourself, you may end up with "a dial tone", which can be ping and online, but does not guarantee the availability of storage space. "

In AWS's SLA for SQL Server systems, it promises 4.5 minutes of downtime per month, or 99.99% uptime. However, like Azure's SLA, the uptime guarantee applies only to two or more database instances that are paired with each other.

Google's cloud platform also promises only 4.5 minutes of database downtime per month. However, downtime is defined differently in Google SLA, which refers to "loss of external connectivity or persistent disk access to all running instances", where instances are placed between two or more zones in the same computing zone when deployed.

Prepare for system outage

Although the cloud is a very useful resource, it is more important for us to realize that cloud services are not absolutely reliable and have the possibility of failure. This is why Bermingham shares SQL Server high availability best practices and what contingency plans should be made in the event of an outage.

Bermingham says that first you need to make sure you have redundant Internet access and a plan for how the application connects to the SQL Server and how the client connects to the application. Accidental loss of Internet services is a single point of failure in which users need to find a response or solution.

In addition, one of the things that users need to consider is the use of multiple availability zones to deploy SQL Server in the cloud. Because available zones may be completely unavailable, Bermingham recommends deploying databases across regions to compensate for complete outages.

Bermingham stressed the importance of implementing disaster recovery plans because data in the cloud can be lost in a variety of ways, and natural disasters such as fires and floods can damage physical servers that hold important data, but most of the disruptions we see today are caused by some kind of human error.

Take advantage of high availability tool

When implementing SQL Server high availability best practices based on cloud systems, it is important to ensure that important data is available regardless of the circumstances.

Bermingham says that even if you are running SQL Server in the cloud, you need to manage it and make sure it is online all the time. The high availability features in SQL Server can also help DBA manage databases in the cloud, such as Always On Availability Groups and Always On Failover Cluster Instances.

In SQL Server 2012, Always On Availability Groups binds a set of primary databases to up to eight corresponding secondary database sets that are configured to fail over together in the event of an outage. Bermingham indicates that because SQL Server runs on each instance, this technology enables very fast failover.

He added that page fixes can be done automatically without using third-party products, and that DBA can back up and run reports on secondary systems. However, Always On Availability Groups protects only the user database, not the system database used to manage and maintain SQL Server.

Always On Failover Cluster Instances uses the features of Microsoft Windows Server Failover Clustering (WSFC) to provide high availability protection for the entire SQL Server instance. Failover cluster instances are deployed across multiple WSFC nodes for redundancy in the event of a failure. Therefore, DBA does not need to manage availability or maintain passwords and usernames in multiple instances.

Each cloud platform provider also provides the ability to manage storage availability in SQL Server systems. For example, Microsoft provides Azure Managed Disks, a software program released in 2007 that reduces potential storage disruptions to Azure VMs by transferring data between different storage units.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.