EIMS: Disaster Recovery Procedure

EIMS Business Continuity

EIMS’ policy is to pre-empt catastrophic business failure by ensuring business continuity. EIMS is a multi-site, multi-national organisation, with a network of distributed offices, people and technology. It is this organisational architecture, designed to have no single point of failure, which is at the heart of our business continuity strategy.

Many details of our data security measures and disaster recovery procedures are confidential, and available on request under NDA. What follows is a broad outline only.

Data Centers

Our business relies on information and communication technology to operate. We operate multiple, independent data centres, each of which is capable of supporting business operations. These data centres are separated by long geographic distances, such that damage to one is extremely unlikely to involve damage to another. Each data center has multiple (between 2 and 6) communications lines to the outside world, such that damage or congestion on any one line will not affect the others. All lines support both voice and data, such that failure of one line will not completely disable any one service. Each line is provided by a different telecommunications provider, and runs over a different route, such that no single JCB digger or explosive device would disable communications.

Each data center contains all the usual high-availability technology available in a modern data centre, including highly available servers, with multiple power supplies, network connections, RAID disk storage, data backup and restore, air conditioning, load-balancing, Uninterruptable power supplies, and in our main data centre, backup generators in case of prolonged power failure.

This means that in the event of complete failure of one data centre, at least one of our data centres will be accessible from each of our operational offices in every circumstance except a massive global fuel shortage or a nuclear holocaust, in which case we’ll probably have other things to worry about than whether or not EIMS is still operational.


All major offices are connected both to the internet and directly to our data centres via multiple leased lines. Lines are provided by multiple different telecoms providers, running over different routes, such that damage to any one line of communication will leave the other operational. Each line is capable of supporting both Voice and Data communications, ensuring that the office has ‘always-on’ access to both our data centers and the internet, Small / back-office operations may use DSL backups in place of leased lines, for reasons of economy.

Operational Centers

In addition to technology, each office has a self-contained complement of management and operational staff, such that it may operate independently for day-to-day operations. This means that campaigns being conducted on behalf of clients can be operated from any customer-facing center, even if another center or head office goes offline. At the time of writing, EIMS operates 3 customer-facing offices and 3 back-office operations which are capable of performing their functions independently for weeks or even months at a time. As project information is stored in the data-centers in a standard format, campaigns can be moved from one office to another, albeit at a reduced capacity in the short-term.


In summary, EIMS has a ‘high-availability’ business architecture. Business continuity is assured by having no single point of failure in terms of technology, premises, people and processes. Should a major disaster such as a bomb or a falling airplane cause the complete failure of any one data center or operational center, the other centers would continue to operate, albeit at a reduced capacity.

© EIMS Ltd. Any re-use of any element of this website, including without limitation all text and graphics, whether in part or in whole, whether modified or not, must acknowledge EIMS with a clearly visible link back to www.eims.biz, such link should be placed in reasonable proximity to the re-used element, and should link as closely as possible back to the original element.