A decade ago, companies like SunGard grew very quickly by selling a solution most of us would call “cold site disaster recovery.” In essence, SunGard compiled a number of physical servers and compute capabilities, then leased these servers to many companies to be used in case of a disaster on a first-come, first-served basis.
SunGard built facilities in cities around the country such as Chicago and Philadelphia where users could converge when a disaster struck their primary data center. By loading their applications onto the DR servers and mainframes, users could attempt to recover their operations from “cold” servers.
These cold servers were bare metal servers – no operating system, no applications, no patches and no data. Once the tape backups and the IT recovery team finally arrive at the recovery site, the system recovery begins the time-consuming process of loading the system from the tapes. And as if that isn’t hard enough, tape backup is notoriously unreliable. Tapes have high failure rates when it comes to trying to read a tape from a different tape reader than it was written from. It’s not unusual for tape readers to need recalibration just to recover the data on the tapes for DR purposes.
The typical cold site would allow for a 48 hour pre-scheduled annual disaster recovery test. Unfortunately, most of the users I’ve spoken with were unable to recover their entire system during the test window due to the time-consuming task of system recovery. This leaves the remaining untested part of the DR plan tied to the “hope and pray” strategy – not a particularly comforting position to be in for a mission critical IT infrastructure.
Cold site DR is dead.
With the increasing demands for 7×24 IT availability, recovery time requirements have dramatically shrunk over the last decade. While 10 years ago, a 3-5 day disaster recovery time was acceptable, most businesses are now demanding 1-4 hour recovery times given their heavy dependency on their digital infrastructure and electronic assets. This requires an entirely new disaster recovery strategy.
Fortunately, fundamental shifts have taken place over the last decade that makes it far easier and more cost-effective to achieve significantly faster disaster recovery times. Faster, more cost-effective Internet connectivity, point-to-point networks, more cost-effective SAN storage and the advent of cloud computing have changed the rules for disaster recovery.
One of the most significant advantages to cloud computing is what the cloud delivers in terms of disaster recovery. Cloud Computing delivers faster recovery times and multi-site availability at a fraction of the cost of conventional cold site disaster recovery.
Cloud computing virtualization delivers a very different approach to disaster recovery. The entire cloud server, including the operating system, applications, patches and data is encapsulated into a single software bundle or virtual server. This entire virtual server can be replicated to an offsite data center and spun up in a disaster recovery cloud in a matter of minutes.
Since cloud servers are hardware independent, the operating system, applications, patches and data are safely and accurately replicated between data centers, removing the burden of reloading each component of the server. This dramatically reduces cloud server recovery times compared to conventional disaster recovery in which physical servers need to be loaded with the OS and application software, then patched to the last configuration used in production – all before the data can be restored.
Cloud-based disaster recovery delivers warm site recovery times more cost-effectively and without the drawbacks of conventional cold site DR approaches. It also enables much faster recovery point objectives (RPOs) – eliminating the risk of data loss when failing over the DR servers.
With cloud-based DR, the server replication is dramatically accelerated and network replication becomes the critical path to recovery, including IP address mapping, firewall rules & VLAN configuration. Solutions like Online Tech’s DR Now! cloud-based disaster recovery not only replicates the servers between data centers, but also replicates the entire network configuration in a way that recovers the network as quickly as the backed up cloud servers.
Cloud-based DR delivers a set of benefits that make it difficult to look back at cold site DR with any sense of nostalgia. Long live cloud-based disaster recovery!