SQL server clustering and replication
One way to help boost your SQL server uptime is by clustering SQL servers.
Clustering can be best described as a technology that automatically allows one physical server to take over the tasks and responsibilities of another physical server that has failed.
Clustering in SQL Server 2000 or 2005 is when two or more servers (also known as nodes) are configured to operate as a single virtual SQL Server to support database transactions.
Should one SQL server fail in the cluster, another clustered server will automatically take over, keeping downtime to a minimum* rather than hours or days as in normal circumstances, and you will notice little, if any differences during this time.
*We predetermine the time lag before the secondary servers comes up online. This ensures there are no false starts. A common reason behind false starts in this situation is the primary server rebooting after patched have been applied.
It’s important to remember that clustering is not a complete backup system for your applications. It is only one part of a multi-part strategy required to ensure minimum downtime and 100% recoverability. (To find out about server backup with Veritas, the fastest disk based recovery on the market, click the link.)
UKFast SQL clustering provides the added advantage of no single point failure and geographical flexibility. Data redundancy is achieved through each server within the cluster holding replicated copies of clustered data on its own local disks. Data is continuously replicated across a shared or private IP LAN or VLAN. As standard, SQL clusters provide protection against server or application failure, with UKFast adding even greater protection against geographic disasters and data redundancy.
When choosing clustering for your dedicated server there are two primary clustering methods:
- Active-Active: both servers in the cluster process the transactions typically at 50% of the total resources. This is because either server can failover to the other so that a single server would be responsible for the transactions of both servers.
- Active-Passive: only one server in the cluster processes the transactions. The second server only processes transactions if the first server fails. To run in this mode the servers require the additional Replication management software called Replic8.
Whilst SQL replication maintains a single copy of the data and provides failover within a single site, UKFast clustering offers added data protection, assuring continual access to data and applications across sites. SQL clustering allows you to run your SQL cluster in "Active-Active" mode which allows both (or all) servers to be always on and there will be zero downtime should there be a problem with any one of them - providing data redundancy.
Our SQL allows clustering with the added advantage of no single point failure and geographical flexibility. The data redundancy is achieved through each server within the cluster holding replicated copies of clustered date on its own local disks. Data is continuously replicated across a shared or private IP LAN or VLAN. As standard, SQL clusters provide protection against server or application failure, with UKFast adding greater protection against against geographic disasters and data redundancy.
Data replication with Replic8
Replic8 uses patented replication and failover technology, which continuously captures byte-level changes as they happen and replicates those changes to one or more target servers at any location.
A small software module is installed on each source and target server. An easy-to-use client application lets you define which data will be replicated and where the copies will be stored. The initial mirroring operation can be performed while the source servers are active. After this simple configuration and initial mirroring are completed, operations are automatic and transparent.
The failover module resides on the target (second) server, and continually monitors the source servers. In the event of a server failure the target server can assume the names and IP addresses of the failed servers (in addition to its original name), and invoke scripts to restart applications.
Benefits of clustering:
- Zero downtime.
- 100% availability.
- Load is shared between all live servers.
- Replicate over any distance to any location.
- Replicate using standard network protocols.
- Minimum bandwidth usage for replication.
- Queued task processing for automatically trigger tasks.
- Microsoft Gold Certified Partner.
Benefits of Replic8:
- Quickly restore access to data and applications.
- High availability and failover.
- Reduced downtime.
- Many-to-one failover.
- Replicate using standard network protocols.
- Minimum bandwidth usage for replication.
- Queued task processing for automatically trigger tasks.
- Windows Certified Partner.
With traditional backup alone, you run the risk of losing up to a full day's data in the event of a system failure, and this could take hours to restore. For many businesses, this amount of lost data and downtime is no longer acceptable, that’s why industry leaders such as Microsoft and IBM use this type of replication for their high availability and disaster recovery solutions.
How SQL server clustering works:
You many be asking yourself what is the difference between clustering and load balancing?
Well, clustering is having several servers doing the same function so load can be shared. This addresses capacity issues. Clustering is usually carried out using database servers.
Load balancing is having the same application running on multiple servers. This benefits redundancy. Essentially load balancing can be a described as a type of cluster. Load balancing is usually carried using other types of servers rather than database servers.
To set up clustering on your SQL servers simply call us now on 0800 4584545, and talk to an Account Manager, or request a quote online.
