I had to laugh when I wrote that title. Given my recent bout with pneumonia, it has some vague medical allusions. But let's not go there.

The real topic of this post is about information. A recent article I read reminded me of the nuances between these three words, and the significant impact they can have on a network, an IT organization, and the individuals that sell telecommunications services into such a context.

Let's take them out of order. Storage is a broad term that has to do with recording information as it is generated. It has to do with the world of Storage Area Networks (SAN) and Network Attached Storage (NAS) and the real time information capacity needs of an organization or even an individual. You think about storage every time you worry that your hard disk is running out of space. For a telecommunications infrastructure, the issues here are about total tolerated delay and total bandwidth required. You see, storage is often replicated to make the information system survivable.

Backup, however, is about information loss. A backup is a snapshot in time of a storage system. You're thinking about backup every time you wonder how much information you are likely to lose if your hard disk fails. For most company, backups are stored offline and off premises. They serve as a fallback in the event the storage system fails. Even if a storage system is replicated (e.g., backup data centers), a backup system is important because it makes it possible to recover the storage system at a particular moment in time. For telecommunications systems, backup is important only if it is being done online to a remote location. Then the network needs to have sufficient capacity to carry the backup load.

Then comes retention. Sounds like backup, doesn't it? But retention is about keeping, and being able to access, specific information for specific periods of time. You are thinking about retention when you consider how many electronic tax returns you should keep on your computer, or how long you should keep an electronic copy of the web class you just created for a particular customer.

Could an organization meet its legal and regulatory information retention obligations by simply keeping all of its backup tapes? Possibly, but unlikely. Think of backup as a rather brute-force approach to copying information. It makes copies based on when information changes. Ask any IT director to find a particular file from a particular date on their backup system and they are likely to groan pretty loudly. In fact, the volume of the groan will be proportionate to the size of the backup. Such a request is something like asking an excavator to locate a particular load of sand on a two-mile stretch of beach they just covered to deal with erosion problems. And, if you check, you find that few organizations want all of their information retained. They want only information they are legally required to retain, or which has a business import.

Retention is not just about making a copy of the information, it's about being able to access specific information in the information set. It's precise - specific - selective. Organizations that have large volumes of critical information need to understand the distinctions between these three information terms, and use the right tools for the right jobs.