It can be easy for those of us who live and breathe MultiValue databases every day to forget the outside world exists. After all, we’re all so busy juggling massive amounts of data and programming requests that we sometimes tune out other important factors that can affect how well IT organizations function. One of the main issues most commonly overlooked is disaster recovery, which is often treated with an “out of sight, out of mind” attitude as we deal with more pressing issues. In fact, high availability/disaster recovery should be a critical part of every MultiValue deployment.
What is HADR?
High availability (HA) is a complete solution able to quickly and accurately recover from a system failure. Usually, the intent is to recover to the same server or to have two servers clustered together to recover quickly for high availability. Disaster recovery (DR) is the ability to recover accurately to a second server, usually at another location. A DR server may be across the room, across the campus, across the city, or across the nation. Should the primary data center burn to the ground, the applications and data would be located at another facility and therefore be preserved. The combination of HA and DR makes it possible to recover information with zero data lost, avoiding the minimal data loss associated with disaster recovery implementations to a remote location. HADR also offers the added bonus of eliminating outage times even for scheduled maintenance and upgrades.
Why HADR Matters to MultiValue Database Users
Of course, having back up is important no matter what kind of computing one is doing, but it’s especially relevant for MultiValue because so much of what we do is in real time (or nearly real time). In practical terms, MultiValue data can’t be archived or backed up once a day – it has to be constantly backed up because even a blip of a few seconds can have a significant impact on information integrity.
But what if your life depended upon it? In fact, for millions of Australians, MultiValue and disaster recovery is making a critical difference in healthcare delivery. Most pharmacies down under rely on this approach to ensure they never lose data. Can you imagine the pressure the pharmacies must face to ensure all drugs are dispensed correctly in the hospital, avoiding any unknown drug interactions? Missing a single prescription or giving out the wrong drug or dosage could be catastrophic for any patient.
And, for more than 20 years, hospitals in Australia have been relying upon MultiValue. But MultiValue alone is not as effective as MultiValue paired with HADR.
The Perfect Combination
HADR and MultiValue together offer an unparalleled combination of modularity and reliability. A MultiValue database stores information according to different “dictionaries” for flexible research and data modeling, and one system could contain multiple effective ways of reading the same piece of data. This creates a much more customizable database, allowing users to set their preferences to suit the needs of their business, whether they work in finance, medicine, industry, or entertainment.
But MultiValue is also much more modular than competing styles, storing each category of data in its own dynamic array. This means data lost is only lost from a single category, avoiding an ugly domino effect as one piece of corrupted information ruins an entire database. Throw HADR into the mix and not only is data loss less catastrophic for other packets, every loss is by nature quarantined. The HADR system guarantees a high rate of recovery, while MultiValue means less data needs to be recovered overall, with less catastrophic results if 100% recovery cannot be achieved.
Like peanut butter and jelly, MultiValue and HADR are a perfect pair!