Disaster management and redundancy planning is an aspect of the industry which is growing and poorly catered for. Initially while I was employed at Tech-Precision we began to develop this area because for the most part as a third party hardware maintenance company we had a significant investment in hot standby spares which produced no income but were a necessity by nature of the business. Customers were reducing their maintenance cover while the uptime requirements were increasing. This combined with the exorbitant charges by the manufacture for disaster services resulted in the fact that generally only a few companies would take up these options.
I prepared a strategic specification for marketing and the implementation of a disaster recovery plan, which was scaleable to the client’s requirements and budget. This covered the very small, minimum requirements, such as items of hardware that were critical to client, through to full remote disaster sites running in a hot standby environment and covered all aspects from internal resources through tasks which were out sourced and generally forgotten in most disaster plans. In the smaller cases it could be as simple as supplying and configuring extra disks, or peripherals for those companies that chose not to invest in more substantial plans.
In the larger cases we would work with the client to evaluate what there needs were. This would generally start with a Requirements Definition identifying the services required and the time frame that was available from the beginning of any disaster up to the point where the above services had to be in production. Generally with most companies we could develop a timing schedule, e.g. not all services were needed within the same given time frame.
For example I implemented a full remote site disaster plan for a xxxxxx company, which administers approximately $53 billion in funds. In there case failure to perform certain tasks as a xxxxxx resulted in heavy fines incrementing on a daily bases. The cost justification for any disaster plan must take into account all costs. Another formidable company who specialises in disaster plans among other things omitted taking into account penalties, out sourced tasks, and the capacity to utilise as much of the investment as possible while not in disaster mode. This combined with my insistence that the plan must be tested prior to any disaster enabled me to secure the business.
As a result the client implemented the full plan which included the supply of the required VAX’s and network components, the development of the remote environment, the backup and restore strategy, and it enabled them to reduce there 18 hours shifts, due the available computing resources while not in disaster mode.
These are some of the points covered in the above example:
Real cost, both capital and ongoing Investment method, rental, purchase
Savings, by way of added performance Potential cost from the lack of any plan
Environment, location, air-conditioning & monitoring Projected expansion required
Resources required to maintain the plan Staff relocation costs, and availability
Security, backup & restore procedures Implementation timing chart
Network routing, relocation of Telcom services Impact of a disaster at the disaster site
Impact of a disaster at out sourced locations
While it is true that I have not lived and breathed disaster management, I doubt that anyone has given the number of sites that are properly prepared. However I have set-up many different sites for differing reasons, along with setting up disaster proposals. While most of the above has revolved around Digital Equipment Corporation hardware, the principles remain the same irrespective of the hardware platform.