This winter I was treated to another family reunion with three or four generations of Toigo kin in attendance. The group included a couple of my favorite uncles, both septuagenarians, who still enjoyed chomping their cigars, drinking their sweet vermouth, and ruminating about their lives in the technology-challenged past.
Uncle #1 worked for NASA in the early days and can actually remember when eight KB of storage occupied a structure about the size of the Vehicle Assembly Building at Kennedy Space Flight Center. He still fumed over the decision to "waste taxpayer money building that Johnson Space Flight Center." He has never changed his position that the Texas space center was "pure pork barrel, since we could run the entire Mercury program from Goddard in Maryland." (Curmudgeons run in my family line, as readers of this column might suspect.)
My other favorite uncle, call him Uncle #2, loves to taunt me with stories of how he, with no formal training in technology, took over the data center of his corporate conglomerate and "fired all those slide ruler-toting, pencil-necked, pony-tailed, propeller heads" (i.e., the IT staff) because they didn't have "a lick of sense about how to run a business."
"Back in the 70s," the codger reiterated his oft-repeated tale, "we had big iron. You needed to watch yer storage like a hawk because adding DASD -- those disk drive units the size of Frigidaire refrigerators -- required a heck of a lot more planning and a lot more
Clouds of smoke from his torpedo swirled around his head as he continued his reverie, "Just lettin' those users suck up storage resources was a sure path to business failure. Every time you added more storage, you had to build a new building, air condition it, and upgrade your UPS, Halon, sprinker systems, physical security -- the works. It cut into profits for the whole darn company."
He recalls that when he discovered that the DP manager had told the systems programmers to go out into the departments and ask them what they might like to have in the way of applications, he simply "fired the whole fool lot of 'em."
"They just didn't get it -- adding more applications and end user storage would eventually drive the company to bankruptcy. That's the problem with you young folks today: no sense of responsibility."
When I posited that maybe the DP manager was just trying to make sure that IT was providing excellent service, perhaps to forestall the coming tide of open systems computing, he looked me square in the eye and squinted. "Is that a propeller I see on your head, boy?" He checked around the back and said, "No. Worse. It's a pony tail."
It was an old saw and I knew it by heart. Everything wrong with IT generally, and data storage in particular, was caused by those "New Age, avocado dip, creamy-smooth, liberal" IT guys who sought to "empower" end users and departments with their own servers. The move, in his view, ended the era of IT discipline and ushered in the era of irresponsible computing.
He talked about a new idea that he was about to try at his shop, hierarchical storage management (HSM). "We figured that matching data based on its usage characteristics to the least expensive storage medium -- optical disc or tape or what have you -- was a good move. You just keep the most frequently used data on DASD, and push everything else down to tape."
"But," he lamented, "We never got the chance to try it out. Wouldn't you know it, the network guys like Datapoint were working in cahoots with those pinkos -- you know who I'm talking about, the Data Generals and Hewlett-Packards and Control Data Corps and Digital Equipments of the world -- giving computers directly to the end users. Heck, they were installing PCs right under our noses and calling them typewriters. It's been a slow slide into perdition ever since. I always knew it would get us into trouble. Look at Enron."
"And Johnson Space Flight Center too," chimed in Uncle #1, pouring himself another Vermouth.
I didn't want to debate the Enron issue, but I saw where Uncle #2 was going. I mentioned to him that times were changing and a number of companies were waking up to the truth about the cost of using a single platform for all data storage. HSM is being resurrected as a strategy as more and more storage is placed into networks or fabrics. Several big name storage vendors have released or are preparing to release low cost IDE/ATA drive-based arrays into the market for use as a secondary storage platform, while a broad range of disc and tape vendors -- many of whom are well known from the days of mainframe HSM -- are re-tooling their marketing messages for the networked storage era.
"It'll never happen," he retorted. "It would require businesses to start treating IT like a business instead of a special entitlement that the company owes to end users. Plus, the vendors who are making all their money selling their high-end arrays would never stand for somebody selling cheaper gear that delivers practically no profit margin. They'll just say that it's more sensible to build a high-end storage utility out of a bunch of high-end arrays that any application can plug into. Ultimately, they want to sell you more of their expensive stuff."
I was going to argue the point with him, but I deferred in the interest of preserving the festive mood. I learned long ago never to discuss politics or data storage at a family function. I must say that Uncle #2, always ready for a good debate, seemed a bit disappointed at that. He casually mentioned to my five year old that her daddy had "long hair like a girl."
Epilogue: In late February, I had the opportunity to moderate a panel discussion about networked storage at an event in New York -- a Marcus Evans conference on Disaster Recovery and Storage Management, which was also covered by SearchStorage. I asked about the possible rejuvenation of HSM in light of the forthcoming release of inexpensive storage platforms from many major storage vendors.
A panelist from IBM, a company with no specific plans to release such a low-cost array, shut me down, "I find it difficult to see how you could make a business case for such a strategy based on the fractional cost savings -- measured in peta-cents [sic] -- that would accrue to migrating data across less expensive arrays."
I realized that curmudgeony Uncle #2 had called it correctly. As he would say in his kinder moments, "Even a broken watch is right twice a day."
About the author: Jon William Toigo has authored hundreds of articles on storage and technology and authors the monthly searchStorage "Toigo's Take on Storage" expert column. He is also a frequent site contributor on the subjects of storage management, disaster recovery and enterprise storage. Toigo has authored a number of storage books, including, The Holy Grail of Data Storage Management.
This was first published in February 2002