This article can also be found in the Premium Editorial Download "Storage magazine: Storage managers reveal details on tape reliability."
Download it now to read this article plus other related content.
Do you find it frustrating to deal with vendors? Why do we use acronyms for things that don't actually mean anything? Is there any tangible human benefit to having "a secure mobile transport for valuables such as currency, identification and automotive control elements with privacy features"? Can't you just call them pants?
I suppose it's good news for guys (like me) who make their living translating vendor babble, but even I'm starting to lose my mind with all the senseless misdirection in business.
So my New Year's resolution is to start translating erroneous and/or misleading "acronym-filled horse*&^%" (AFHS) into plain, simple English. (And there's no disrespect intended to my non-English-speaking friends because I can barely speak my own language.)
Let's begin with virtualization, potentially the mother of all meaningless storage terms. Simply put, virtualization is the abstraction of the physical storage device(s) from the application or operating system layers. In the most-used example, it's a volume manager, whereby the operating system and applications think they see a physical disk but what they're really seeing is a virtual representation of a disk. Who cares? No one.
The "virtual" factor isn't the issue (we've been doing it for a zillion years); the buzz is around where it specifically executes. Volume manager runs on the server. RAID virtualization runs on arrays. The "fabric" itself is about to execute these and many other functions,
Information lifecycle management (ILM) is the same deal. It's not a product but a philosophy about what goes where, when, and under what conditions and changes. Today you care about tiered storage, which is a way of asking, "How can I use my cheap stuff more effectively along with my expensive stuff?"
Wide Area File System (WAFS) is a brand-new buzzword, which I promise will be the confusing term for the next six months. What it means is the ability to use CIFS and NFS file-serving protocols effectively over a wide area network. We all hate remote IT. We hate having people, gear and, most importantly, data outside of our core IT operations where we have some hope of actually protecting, managing and mining that valuable stuff.
So forget WAFS--it's just another silly acronym. What it really means is consolidating geographically distributed data services, otherwise known as GDDS. Oops, I did it myself!
Anyhow, there are several new, cool approaches to eliminating remote office file servers, tape stuff, backup/recovery and general IT staffing. Vendors such as Tacit and Actona Technologies (acquired by Cisco) actually enable a central IT server to kill the latency problems associated with distance, so the remotes don't need anything other than a glorified cache locally--which is way cheap and has tons of promise. Other companies like Riverbed Technology and Signiant accomplish the task by compressing the data that traverses the wire, and eliminating duplicate data blocks so only unique data has to move. Both methods save tons of money that's wasted on inefficient telco lines, and will radically improve overall data availability and recoverability.
My real problem is that all the AFHS--when you rip out the garf and tell me what it really is--is usually pretty cool. We need to nuke remote IT. We need to virtualize. We need more efficient and economic storage infrastructure plays. So vendors, do us all a favor and just say what you mean.
This was first published in February 2005