Foreword
Every new version of SQL Server brings with it new tools and features for database
administrators (DBAs), developers, and architects, for them to be able to provide
an effective solution for the end users in a simpler and more efcient manner. The
terms effective and efciency can be measured in a technical perspective as High
Availability (HA) and Disaster Recovery (DR).
Let's assume that a world-wide retailer's CTO has been tasked to provide high
availability for their mission-critical back-end systems that are built upon SQL
Server technologies. It is a major task for DBAs to implement an HA solution on a
SQL Server data platform, and for developers to ensure that underlying application
architecture is HA-aware to take advantage of mission-critical features from SQL
Server technologies.
SQL Server High Availability features include failover clustering, database
mirroring, log shipping, replication, and backup and restore strategies. The solution
must t within the company's budget, keeping up the scalability, and should reduce
degradation in performance. When it comes to the disaster recovery solution, it may
not be possible for the primary and secondary site's hardware to be the same, such
as fewer hardware resources, limited storage capacity, or data center limitations. It
may be acceptable to the business in terms of costing, but when a disaster strikes, the
damage will be irrecoverable. All such out-of-the-box limitations must be avoided, or
reduced, to keep up the mission-critical applications always online.
The most critical aspect of any high-availability solution is designing a failback
(quick recovery of data) strategy, that is, the application must be designed to direct
the connections to the available instance when the SQL Server instance fails over to
the secondary server in a seamless manner. Though SQL Server is an out-of-the-box
disaster recovery tool, many of them are limited to traditional methods of backup
and restore functionality that are time consuming during a recovery event. So,
the new trend in the industry is to replace the existing older backup and restore
policies, taking advantage of hardware capabilities as a DR solution. This requires
server-class hardware and standardization on same hardware, but still lacks the
quicker data recovery strategy.
评论0