Last week, we examined the differences between a disaster recovery hot site and a remote site. While there are distinct differences between the two, both hot and remote sites must have multiple levels of redundancy designed and built into every aspect of the facility.
Below is a quick DR infrastructure checklist to help in your planning:
Ensure your DR provider has redundant network equipment
Consider using multiple network providers; Some colocation facilities have over 30 network providers for maximum redundancy
There should be multiple sources, ideally sourced from different power grids
Are there backup power generators?
Is there onsite fuel to run those generators? You’ll want onsite fuel that can last a few weeks.
Servers and other systems generate a significant amount of heat, making backup cooling systems a key component of a disaster recovery facility
For data and telecommunications, your disaster recovery partner should deploy an uncompromisingly high level of security through technologies such as:
--Virtual private networks (VPNs)
--Virtual local area networks (LANs)
--Firewalls and more
Physical security is also important. Ensure the provider has 24-hour manned security at the DR facility as well as cameras and digital monitoring.
Whether it uses servers, routers or T1 lines, your remote or hot site provider should have “N+1” availability, a system configuration in which multiple components have at least one independent backup component to ensure system functionality continues in the event of a system failure
The best deployments use Redundant Array of Independent Disks (RAID) methodologies to “stripe” data across systems for performance and data mirroring for improved protections and availability. Striping means that all available hard drives are combined into a single, large virtual file system, with the blocks of the file system arrayed so that they are spread evenly across all the drives
Ideally, your remote-site provider can accommodate multiple strategies, including redundancy, clustering, load balancing and warm standby (in which the application is loaded, but not running).
We know a lot going into implementing and maintaining a disaster recovery system, so we created this handy Disaster Recovery Guide.
You can also read our other posts on disaster recovery here: http://www.eci.com/blog/categories/Disaster_Recovery.html
Finally, contact us to speak with one our experts about our Eze Disaster Recovery Services.