Last time I mentioned about Contingency Plan. Today I further the discussion of it in a practical sense and I would bring up something about Disaster Recovery (DR) which is actually a kind of contingency plans.
The need for DR is to ensure the business continuity whenever the crisis arises such as fire, power failure, storm, disease outbreak (e.g. SARS) and any other unexpected events which can damage your business, and your precious data.
Smit (2007) reports that:
‘According to the Meta Group, the average cost of an hour of downtime for data centre applications is $330,000. According to the Strategic Research Corp., if that data centre belongs to a credit card authorization company, the loss jumps to $2.6 million. And if it belongs to a brokerage house, it climbs to $6.5 million. One day of lost productivity costs a company an average of $432 per employee.’
Without doubt this is a great loss to a company. Don’t expect your clients would understand your difficulties and accept your apologies. The best solution is to plan ahead before the disaster occurred. Reducing the downtime means cutting down the loss. But how?
Ibid (2007) has given us the directions to ensure high availability and business continuance.
Protecting, replicating and backing up data
First of all, we need to build up a high-capacity and low-latency data centre, which is interconnected to MAN (Metropolitan Area Network) and WAN (Wide Area Network). This can enable zero-data-loss data mirroring to protect user sessions, prevent transaction loss, and support automatic failovers between mirrored sites. SAN (Storage Area Network) technologies which enhance the distance, security, bandwidth utilization of replication and backup to remote sites, however it has not been really popular. In addition, technologies such as write acceleration, tape acceleration and server-less backup reduce latencies, extend distances and reduce application impact of storage replication applications. Moreover, it needs support for business continuance applications, especially those that provide replication and data protection.
sourced from Javvin
Enhancing application resilience
Companies can remove single points of server failure by deploying high-availability clusters or load-balancing technology across Web and application servers. Apart from that, connectivity can be extended between clusters in different data centres to protect against major disruptions. Achieving this type of redundancy requires a high-speed, low-latency metro network.
Ensuring user access
Companies can employ technologies such as VPN to allow users from branch offices and telecommuters to reconnect to applications quickly as soon as they are up and running. In addition, technologies such as global site selectors can allow users to manually or automatically connect to the most available web application available at any given time. In the case of a disruption in any one application environment, users continue to have access to the alternate site.
Needless to say, we all realised the devastating impact of 911. It just happened once in the past 6 years. Do we really have to focus on this incident too much and then, spend tens and thousands dollar on the above systems. Some may not be used even once in 10 years. The answer is absolutely. I still remember the disaster happened around 6 years ago. Due to the disorder of the fire sprinkles of an office on the high floor, it flooded the whole commercial building with water and thereafter, the power was suspended for a day. At that time, what we could do was to shut down all our mission critical servers before the UPS (Uninterrupted Power Supply) has been worn out. This action was to protect our servers and data. Lucky we had installed the UPS for all mission critical servers.
You can probably imagine how big the loss was caused by this incident. Very unlikely, DR is able to fully eliminate the loss but at least, it can lighten it. Anyway, DR is totally a choice of investment. What is your choice?
To be continued
References
Javvin, ‘Metropolitan Area Network and MAN Protocols’, Javvin Technologies, Inc, California, <http://www.javvin.com/protocolMAN.html>.
Javvin, ‘Storage Area Network and SAN Protocols’, Javvin Technologies, Inc, California, <http://www.javvin.com/protocolSAN.html>.
Javvin, ‘WAN: Wide Area Network’, Javvin Technologies, Inc, California, <http://www.javvin.com/networkingterms/WAN.html>.
Smit A 2007,’Data centre safety measures protect business’, Enterprise Innovation, Technology, posted 28 August 2007, viewed 8 September 2007, <http://www.enterpriseinnovation.net/article.php?cat1=2&id=1847>.
No comments:
Post a Comment