Fault tolerant architectures are the very life blood of any company. Lose your IT infrastructure and/or your data and you could very well lose your company. You might be skeptical about that previous statement but as a “for instance” check Gartner’s statistics regarding the number of businesses that did not survive the first World Trade Center bombing because they had no DR plan; it’s shocking.
So today I’m posting a primer on the importance of HA and DR. It contains some industry common language and terms that will enable a better understanding of those terms as they pertain to HA and DR. It will discuss why HA is important and illustrate some of the events which constitute “Denial of Service.” When most people think about losing a data center they commonly think in terms of natural disasters; but what about planned outages such as software upgrades (applications, O/S and etc that shut down the system) or user error that causes data integrity problems and on and on. Want a more concrete example? Consider this quote from The Data Center Journal “The National Fire Protection Association’s (NFPA) book, published in July 2001, “Special Analysis Package, Computer Equipment and Computer Areas” by John R. Hall, Jr., indicates that between the years 1994 and 1998, an average of 1,042 computer facilities fires were reported each year. These fires were defined as “structure fires starting with electronic equipment”.
1042 fires? That’s just under 3 per day for the year! And that’s only one of many factors that can affect the ability of a company to operate their data center.
I live in Florida where Hurricanes (no not the University of Miami Hurricanes ) typically cause problems between June 1st and November 30th (what we Floridians call Hurricane Season). Consider if you will that you are an IT professional working in Florida and have just survived a Hurricane and you and your family are all safe and sound. Typically, after an event like this there will be no traffic lights, no power, no grocery stores, no gas stations and no promises of when those services will return. Are you going to leave your family unprotected and go to work? Probably not and more often than not you won’t be able to.
So what I am driving at is that there are many, many different components and aspects of losing IT continuity and not all of them are just the fire, storm or whatever.
I have attached a PDF file that examines some of the things that you should be thinking about before something happens and how to be prepared for it and to that end I’ve included a primer on how to set up a DR failover architecture for HANA. So if you lose your data center you can continue to support your companies IT requirements from you DR site (which should be sufficiently far enough away to not be affected by the event that took out your primary data center).
Please read and enjoy and by all means ask questions!