0.035: hitachi drops another shoe (it sounded like a slipper!)

I guess I should be ashamed. Embarrassed at least.

Yesterday, someone asked me why I hadn't yet commented here about Hitachi's baby USP-VM announcement made earlier this week, and I responded "what announcement?"

Yep. I missed it. Completely.

Not that I wasn't aware it was coming...Mr. T doesn't make the rounds on Wall Street on the arms of HDS PR folks for nothing, nor do you wake poor old Claus from his year-long blogging hiatus unless something really important is about to happen.

As I had prior blognosticated, there were (and still are) gaping holes in the initial USPV announcement - clearly the team back in Japan had much more in mind than had been announced back in May. And at least a few rumors indicated that many of these would make it through the QA gauntlet before the end of Q3'07.

Guess this stuff really is hard for everyone.

So I guess because I'd expected so much more, the actual announcement of the USPV "mini-me" slipped right under my radar. I probably shouldn't have taken Nigel's advice and stopped stalking HHSNBN (although it has been therapeutic not to read about yet another world calamity that UVM can solve for a couple of weeks, mind you).

But now that mini-me has been brought to my attention, you just know I have to comment. Interestingly though, this time I'm not alone in correcting the typical misleading marketing that the Hitachi Data Systems Santa Clara marketing miscreants have become known for.

It seems I have some new friends in my quest to keep things honest.
 

my helpers heisman the hitachi hype

As with any Hitachi announcement, this one comes with more than its fair share of misleading assertions and what I've come to refer to as "Hitachi Math" - a modernistic form of algebra that arrives at irreproducible results that also have the unique property of having absolutely no bearing on reality. You know what I mean here - nosebleed claims that sound (in)credible out of their sheer ginormity, but have no practical application to what anyone could actually achieve with their kit.

My (newfound) friend BarryW over at IBM dismantles perhaps the most incredulous claim in his recent blog - that there's simply not enough IOPS in mini-me (or big-me, for that matter) to support the amount of virtualized capacity claimed. And I'll extend his argument - not only is there no way to get any useful work done to the purported 96PB of storage, each of the external LUNs would have to be about 6TB in size. Most customer I've spoken with NEVER deal with LUNs that large - and given that Hitachi doesn't support the use of thin provisioning on external storage (until the fall of 2007, they now claim), there's simply no practical use of such large external devices. (And note that "fall 2007" ends on December 22nd, if you'd like to set yourself a reminder).

Now, I'm sure the underlying intent of this Hitachi Math is to imply that since mini-me is designed to that it COULD support such a huge number, it can readily support any smaller amount of external storage.

Which would be an interesting conclusion, except that Chris Evans (The Storage Architect) notes that his hands-on experience with the prior generation mini-me equivalent was that the NSC55 didn't scale very well. He also notes the oxymoronic use of "enterprise modular" to describe mini-me, even though it still suffers from the shared-function failure modes typical of modular (and atypical of "enterprise"). So you might want to ask for proof before you recommend such a solution to your management.

And I can extend Chris' point as well - virtually every true "modular" array includes native support for low-cost SATA storage, yet mini-me remains a fibre channel-only storage device. In fact, even Hitachi's own mid-tier AMS arrays support xATA drives, but not the new mini-me. I think Chris is right in his observation that the USP-VM is no more a mid-range killer than the NSC55, much less the USPV itself. Heck, even ole Claus positioned mini-me directly against the DMX-4 950 and the IBM DS6800 (apparently Claus missed the fact that IBM effectively withdrew the DS6800 from market while he was off doing his Rip Van Winkle thing).
 

de-mystifying migration mania

Another spotlight of this mini-me announcement was an assertion that suddenly Hitachi has cured the world of migration disruptions, complete with broad-sweeping assertions that they can now migrate data off of an old array into a mini-me or a big-me without ANY disruption or risk.

Big claims.

But no explanations of how they are suddenly able to overcome the host-storage WWN/LUN binding without disrupting applications or requiring reboots. Those of us who understand how LUN binding and fabric logins work know that we're talking nothing short of magic here. A host with LUNs bound to an existing storage array simply isn't going to magically start redirecting its I/O to a LUN on a different array (or virtualization devices, such as Invista or SVC) without some assistance. Nor can you pull out an existing USP and fail over to a a new USP-V without at least some disruption.

At least, Hitachi hasn't explained HOW they've implemented this magic. Given the significance, you'd expect an engineer to explain it all, right there in the press release. But no, the quote from (so-called) chief scientist Claus is flavored with more marketing-speak than explanation, leaving yours truly to believe that the marketing guys have overstated the technical abilities of the product.

Again.

And for the record, host-resident solutions are only a part of EMC's migration toolkit. In fact, if you want to migrate with the absolute minimum of disruption, Open Replicator for Symmetrix installs, operates and effects migrations into the Symmetrix the exact same was as UVM, only it's faster and easier than UVM. On top of that, PowerPath users can install a non-disruptive plug-in (PPME), and thenceforth they can in fact migrate entirely without disruption to the host or applications - AND without using host cycles. PPME (PowerPath Migration Enabler) actually implements the magic - allowing the LUN to move and be renamed without disruption to the host or applications. But PPME it doesn't do the copy - that's handled by the array. Nor is it required - you'll just have to take the same disruption as you do with UVM without it.

Now, I haven't seen any claims from anyone that Hitachi has invented similar magic (I'm sure you'll tell me if you have, right?)
 

then there are the togus analysts

I'm not quite sure why, but every Hitachi announcement seems riddled and laced with comments from various industry analysts, many of which no-one has ever heard of until this press release comes out. There's usually one or two Big Names, surrounded by a bevy of analysts who are operating under their own shingle. No offense intended, but I've often wondered why more of the analysts quoted in Hitachi press releases aren't quoted repeatedly, in subsequent press releases.

And since I have met many of these analysts in my career, I'm sometimes surprised at what they say - sometimes, that is.

Most I can write it off as what it really is: nothing more than a paid endorsement, like getting Mr. T to visit Wall Street (I mean, the analysts have to feed their families, too). And within these, there are times when it's pretty clear to me that the analyst has been misquoted, or at least quoted out of context. I know of several analysts who now insist on "final approval" for all quotes, just because some last-minute "copy edit" twisted their intent a little more than was comfortable. You'd think we wouldn't have to resort to "trust-but-verify" in the real world, but hey - it is what it is.

Then there are those analysts who have simply fallen in love with the technology or the company, for whatever reason. Usually the ones in this camp have clout - an established track record and following sufficient to give their word resonance. Tony Asaro falls into this camp, and he wears his passion for large scale, simplified and (ostensibly) virtualized storage in a manner that is legendary. His observations and perspectives are always insightful, but I have to admit I cringe to see him shilling for Hitachi every time he does it. Somehow I just wish he'd raise himself above the vendor wars and keep his focus on a higher plane. But mostly I just hope he and Steve Duplessie forgive me for pouncing their fun with the words ginormous and togus (respectively) this week.

Finally there's the group of analysts and blogging pundits that just seem to have it out for EMC, no matter what, when or why. I won't call them out by name, but we all know who they are. Some I've even dined and exchanged tokens of gratitude with (hey - did you even bring that signed and mounted DMX backplane back to Germany with you?). Most of these folks are so distantly removed from and ill-informed about the realities of today's high-end storage products as to be of no real threat. But their incessant attacks on EMC can be wearing.

There. I feel better - and I don't think I've publicly insulted anyone, identifiably at least (but you DO know who you are, I'm sure).

 

desperate cries for relevance?

I've said it before - Hitachi's marketing madness, with Mr. T, Hitachi Math and the constant drone of "virtualization belongs in the controller" - it all seems to be a latent cry for relevance in a world were customers are trying to consolidate their vendor relationships down to the select few that can provide the broadest, most robust and most dependable set of solutions to help make IT more efficient and a more valuable asset.

I often ponder just how much longer Hitachi can survive being a one-trick storage vendor.

I mean, seriously: where Hitachi has a few engineers that think about storage security, EMC has an entire security Division, with products in use daily by virtually everyone in the world with a bank account, credit card or a remote/home office - and that's just one example. From the market leadership position of Symmetrix, EMC has expanded its portfolio into virtually every conceivable adjacent market, and they've even created a number of new markets outright.

And that's not to mention VMware, which nobody understood until it became EMC's second Wall Street phenom. And even Hitachi can't resist hitching up to THAT pony (thanks for the free advertising!)

All of Hitachi's hype about virtualization, innovation and aging architectures simply isn't passing the most important test of all: Symmetrix remains inarguably #1 in the high end market (and by a handy margin, thank you), serving the performance, availability, scalability, security, tiering and business continuity needs of the most demanding enterprises in the world.

So like Chuck Hollis recently noted, it truly seems like desperate times for at least some of EMC's competitors.

Comments

[Barry Whyte]

I'm glad its not just me! And thanks to you and Chris for pointing out the capacity argument too. While I'm sure their marketing folks are loving it, I'm glad those in the know see it for what it is... the droppings of a large male horned cow ;-)

[Nigel Poulton]

Hi Barry,

May be Im missing something obvious but what, or should I say who, is HHSNBN??

[the storage anarchist]

He Hu Shall Not Be Named.