« March 2006 |
Main
| May 2006 »
I remember in 1986ish, while attending something at EMC's brand spanking new two-story building in Hopkinton, MA (the first of many), Dick Egan said "in 5 years there will be no need for disk drives". He knew that all storage would be put on RAM.
I believed him. I don't anymore. Not even the brilliant Dick, as I affectionately refer to him as, is right all the time. In his case, thank god it didn't - if it weren't for Symmetrix, the poor guy might have only become Ambassador to Guam instead of Ireland. The economics vs. capacity demand never worked out.
I think you may say the same thing today about tape. I'm going to be that the result will be the same - you'll never hit the right cost vs. capacity metric to totally obviate the need for tape. Even more, once I show you how to stop making tapes that contain 98% of the same data each time, you'll be able to spin way fewer tapes, hence saving a boatload of money.
Dave Hitz, super genius and generally entertaining guy, wrote about the potential of Flash instead of disks in this blog - interesting idea. Dave's point is the same as mine, the economics won't bear it out - but his other point is "who cares"? I like Dave.
So never say never my friends. Pretty soon it will all be holographic anyhow....
I didn't attend, but Tony Prigmore, Tony Asaro, Brian Garrett, and Mark Bowker of ESG did. Some interesting things:
1. Dave Dewalt, ESG (not me) Pres said the two companies that keep him up at night are Oracle - because they could get into his businesses and have the muscle and customer base to make a run at it, and Google. Google shows up a LOT in answers to this kind of question - all over the IT landscape.
2. CEO/Pres/COB Joe Tucci was strikingly candid, and that's one of the things I've always loved about the guy. The only other CEO's I've seen as open and candid have been Dan Warmenhoven (NTAP) and John Swainson (CA). Apparently analysts were harping on Joe's $30M pay package, which is sort of stupid since it's all performance loaded - on the stock, which hasn't moved in substance for a long time. If I were a shareholder, which I'm not, I'd be darned happy to pay him a truckload to move the stock value up. I love the stupidity of it all - in the beginning of Joe's tenure, the company was roughly $6B in revenue (and lost money) - and was pretty much a one product company. Now they are about $11B in revenue, making piles of money, have divested the product offering enormously - and the market cap of the company is about the same. This is why I don't trade stocks.
3. Every business is kicking butt, except Legato. I think the main backup numbers were hit, but the ancillary stuff like Xtender and Replistor weren't.
4. Tons of talk about security - finally. It will be interesting to see how EMC plays this out. They have a lot of access points to enter due to all their offerings, so putting it altogether will make for a good story. Dennis Hoffman, who is sort of the EMC security guy, is super smart. As he put it, "security is an attribute of information management".
Hugely well attended, apparently - 4000+ EMC users. Say what you will about these guys, but they have their act together. This was a first class event from everyone I spoke too about it.
Oh, perhaps the best news from ETS was the press release where EMC basically took credit for the fact that the Red Sox won the world series...
The Scott McNealy era finally ended yesterday. Jonathan Schwartz is the new big cheese at Sun. I don't know if that's good or bad yet.
I do know a few things, however. I know that Mr. McNealy is in a very small circle of people who made a difference in technology. Sun, like em or hate em, has mattered. Not the way DG mattered - for a shining moment or two and then for decades of death, but really mattered. Sun made Unix mainstream. McNealy made Sun. Sure he had lots of talented help, plenty of luck, and a great golf game, but mostly, he made Sun.
I also know he rode the bus too long. A smart man (Joe Tucci) once said to me "the thing that makes a zealot successful is what inevitably brings him down". Certainly true in Mr. McNealy's case. Scott fought the wrong battles for too long a time in too public a forum when there was nothing to gain. The Microsoft bashing, albeit entertaining, stopped being smart about 8 years ago. Calling Wall St. analysts sniveling rich punks (my phrase, not his) was seen by some as brave, but most as silly. The fact that he was correct on both counts is irrelevant, unfortunately.
So Sun may have much better stuff than Microsoft - but who cares? Most Wall St. analysts may not be competent to wipe out the urinals at the San Jose public hockey arena, but who cares? It doesn't matter. What mattered is Mr. McNealy forgot the battle that should have been fought - the battle for the customer's dollar and the battle for shareholders interests. Once that happened, there was no turning back.
So, will Mr. Schwartz, the pony tailed wunderkind be the guy to woo Wall St. back? I don't know. The early returns say no, but the street is fickle - a few good growth quarters and all is forgiven. Will he be able to build products that the world wants "today"? Will he be able to make the storage business an actual business? Who knows? But I do know this - Sun, the company, has a customer base to die for, technology out the wazoo, and the cash and mass to make a go of it. Digital should have taught Sun and everyone else a lesson however, that having the best stuff is entirely secondary. If you can't sell it - if you can't tell the customer a smart, cohesive story, if you can't even find the deal so that you can participate - then it doesn't matter how much faster your processor is or how much better your OS is, you will still fail. The bigger you are, the more slowly and painfully you tend to fail - and that is awful to watch.
Can it be saved? Sure it can. IBM was failing and they don't come much bigger than them. Look at them now - completely kicking butt in most every aspect of their business. It was painful, but it worked. IBM stopped doing things just because that "used" to be the way we won - and instead re-evaluated all the assumptions it had made. That's hard to do, but if you are honest with yourself, have a reasonable strategy, and can keep smart people from jumping ship, anything is doable.
Unfortunately, history is littered with those who could have, but didn't. There are very few who look into the abyss and somehow are able to come back. EMC is another one - they were ripe for death at two points I remember - 1989 and 2000. Both times they were vulnerable and in turmoil, and both times they came back stronger than ever. Part of that is because the competition fell asleep and didn't kill them while they could, but the bigger part is because they were honest about where they were, what they had to do, and were committed to doing it.
Roger Marino, the M of EMC told me this as a much younger man - and I never forgot it. When I asked him to tell me the one single management philosophy that he could point to for EMC's success, he didn't hesitate a second: "get a mediocre plan, and execute it violently". Words to live by.
So, in closing, Mr. McNealy - I wish you well. Go teach a class at Stanford. Go surfing. Relax for a change, you clearly deserve it. (I'm not just saying that because I once played a par-3 with him at some event at the Rancho Mirage Marriott and after putting my tee-shot 3 feet from the hole, bravely bet him a million dollars he couldn't get inside of my shot, which he immediately proceeded to do. We never shook hands, so therefore I am off the hook on a technicality. I never learn, however, and did the same thing, for the same result, a few months later with Dan Warmenhoven of Netapp. Same technicality.)
Did you read Tony Asaro's latest? In his blog he talks about the end of SAN and NAS - and he's right. Do you think anyone really cares about SAN or NAS? If you aren't in the storage business, it's a good bet that you don't.
We are way too hung up on what we call things, instead of what they do - or better yet, what we need them to do. No one wants a Storage Area Network - they want a way to have a giant pile of disks be centralized from a management perspective that can act like a ton of direct attached storage mini-piles. The rationale is sound - if we pile them all together we can have them managed by less people and (in theory) we can get better utilization - i.e. better return on assets, out of one big pile vs. a zillion little piles. Do I really care if it's a fibre channel block storage network or an Ethernet based file system interface? Again, not unless you are in this business.
Tony's point, is that you shouldn't need to care - the storage should be smart enough to feed the application requests in the most efficient means possible, and you shouldn't have a blessed thing to do with it. The machine is smarter than you, at certain things. This should be one of them.
It wasn't long ago when part of the system admin's job was to assign physical memory to an application. You assigned page and swap space too. When is the last time you did that? Exactly - the machine is much better at making those decisions than you are. Your job is to figure out what application will lend the most value to whatever area of the business you are in - and then to establish the most effective means of delivering that application value to your consumer - the employee. Your value shouldn't be deciding if it's NAS or SAN any more than it was to figure out the optimal size page file for All-In-One.
In a lot of ways, the storage industry is the most back-assed of all. We have more high-tech stuff jammed into higher-tech packages that forgot the most important lesson of all - how to take the poor human out of the process. Imagine getting in your brand new Infiniti MX-35 - a high-tech car beyond compare. It has every conceivable gizmo, from a bluetooth cell phone connection that automatically routes calls thru the sound system, to a navigation system you can view in 3D. Now imagine that each time you got into it, it asked you to calculate exactly how much gas you would need to get home, based on today's price, once you manually figured out exactly how far you had to go. It would make the sexy new high-tech vehicle seem like a giant pain in the you know what. That's kind of what storage is.
We'll really be on to something when we stop asking if you need 4Gb/second or 2, and start asking "what are your objectives?" and then letting the machines figure out the best way to attain them.
Here's this weeks ComputerWorld article where I complain about followers.
Here's my Computerworld article trying to explain that if we only adjusted our assumptions and expectations, we might actually get things done.
I'm in Orlando with my 4 kids, one wife, one niece, and two in-laws so i won't be blogging this week. I picked Orlando because it's so secluded during school vacation week. You'll probably hear a lot from me next week, as i'm guessing i'll be bolting to the office as soon as I can.
I've got to resign myself that I'm cursed with longer eyeballs than most. It seems I'm like a B movie psychic when it comes to IT sometimes - I see whats going to happen, but there's not much I can do about it.
It's not that I'm smart, mind you. It's only that I have the luxury of watching things for a living. Sometimes if you look long enough, you just see the way the game is going to play out. Security is one of those things.
I'm not the guru - I was just smart enough to hire the guru three years ago (Jon Oltsik). Last year I hired another one (Eric Ogren) because the privacy laws started to be violated. You don't have to be a NASA engineer to figure out that consumer privacy breaches (or threats of breaches - as most of the laws cover) aren't a new phenomenon - only newly reported. I knew it would get juicy.
So here we are - now every day it seems some new debacle is on the news around who lost what consumer data. It's going to get worse - much, much worse. Most states in the U.S. still don't even have privacy laws.
Check out this free abstract of Jon Oltsik's "Protecting Confidential Data" - it's loaded with data that should scare the heck out of you.
The good news is the storage folks - who hold all this data that keeps getting breached one way or another, can no longer ignore the issue, or make nothing statements like "we're going to adhere to any security practice the customer chooses". Protecting data in flight is nice - but only 2% of our data is ever in flight. Protecting the perimeter is nice - but I can't even keep Spam out. Identity management is brilliant - if you use it. I don't care what anybody says - sooner or later you will HAVE to encrypt anything you care about - everywhere it lives.
Here's an excerpt from Hu Yoshida of HDS fame in one of his blogs:
"While the accumulating reports of data loss, have captured the headlines and focused attention on encryption, there is much more to storage security than encryption. Data does not have to be lost to be exposed. A hacker can access storage and steal information, without leaving any trace. Other areas of concern are authentication, authorization, immutability, non repudiation, integrity, privacy, logging, and auditing."
Good point, and as Oltsik will concur - most attacks are from within. Those attacks don't make the Wall St. Journal though - but some tape jockey dropping a tape behind the rack during an audit does.
I guess what strikes me as the most odd in all of this is that the CEO has absolutely no idea (typically) of the sheer magnitude of the potential for problems. They don't know that there are tons of people with "root" privileges in their company that can see anything they want, any time they want. 99% of the time those folks are morally sound and doing all the right things, but sometimes.....
I was in D.C. today, speaking at Softek's sales meeting. To me Softek embodies the concept of boring - they solve real, boring problems and get paid for it. They make money. How boring.
Data Migration is like the dirty little secret of IT - everybody does, but nobody talks about it. I'm not sure why. Big shops have to migrate data all the time from one array to another. They do it because they need newer, faster, better arrays. They do it because their old arrays are off lease. They do it because they need more space. They do it because they want to "tier" their infrastructure. They do it because some sales guy talked them into doing it.
They do it all the time. The average migration in a fortune 500 shop takes about 3 months of planning - and days of praying. Softek moves data non-disruptively - from anything to anything. Softek has about 800 mainframe customers - the kind where "let's hope it works" just won't cut it.
I think that boring old Softek is the perfect kind of company to start to preach the real business value of the solutions they offer - like improved return on assets (the non-disruptive nature probably means 5-10% increased application availability - which means my $50,000,000 IT world just got a much higher return), improved lease utilization (big shops role out arrays off lease seemingly daily, which means they tend to either de-commission them months early, wasting money, or months late, wasting even more money), and a seriously improved return on human assets (not planning for 3 months and not staying up all weekend during a move would probably yield a better life to boot).
These are the kind of boring solutions to real problems that don't get talked about enough - because most folks don't know you can do this kind of stuff. Softek has solved the problem on mainframes, but their own customers open systems world typically doesn't know they also have the answer to their migration issues.
Plus, Sunday night I went to dinner in Georgetown with my wife and some family, and as we're sitting there two big black Yukon's whip up to the front and after unloading all the secret service folks, who hops out but Condoleezza Rice. Impressive, nice lady. Very friendly to the crowd. As an embarrassed republican, it was nice to see a strong, intelligent person such as her to remind me again of why I chose my particular political affiliation.
Check out my latest Computerworld article to find out my thoughts.
Read Jon Oltsik's blog on all the money that not encrypting backup tapes is really costing folks.
Very quietly (I mean, who would want to be noisy about the fact that they need encryption?) the really big banks are spending tons of time and money making sure their tapes are getting encrypted, and the stuff they transfer to partners like credit bureaus is also encrypted.
Sure, they can afford it - but can the rest of us afford not too? I'll be buying laptop encryption this weekend, because my Rolodex is pretty loaded......
Finally, Dave Hitz wrote a blog yesterday about where it makes sense to encrypt disk (versus tape), and while he is correct practically - the reality is that there is no data that matters that shouldn't be encrypted - no matter where it resides. I realize there are practical ramifications to this, but as far as the overriding objective - plan on encrypting everything - eventually.
I started very tired. 98 meetings the first day will do that to you. Interesting stuff included:
Neopath - the file virtualization player is having field success with Netapp and will announce a mongo big strategic investment very soon, and it will turn heads.
Sanrad - no one knows of them, but they have put more "storage intelligence" into more IP networks than everyone else combined. They have a cool box that lets users add an iSCSI front-end to existing or new storage, and provide tons of high-value services, from volume management to snapshots to replication.
Mark Canepa of Sun spent an hour drawing like a possessed Van Gogh while verbally outlining the big picture for Sun storage. The story is good, the question is how many others can tell it.
Had a nice dinner with Rick Belluzzo, big cheese of Quantum. Rick has some outstanding stories regarding Microsoft where he served as President for a few years at the turn of the century. Fascinating to get real inside perspective on such a mega-machine.
Had a nice bottle of wine with my old pal Nancy Hurley, but can't for the life of me remember what it was.
My lunchtime presentation sucked. Normally there is enough time for folks to eat before I talk, but this time it was to cramped, so i had to speak while plates and glasses were clinking all over the place. I couldn't hear myself, so I'm guessing plenty of others had issues as well. Hopefully, this issue will be fixed for the fall show. Plus, I screwed up (though I blame Lotus Notes) and sent the wrong presentation, which i didn't realize until i stepped on stage.
What was interesting was the sponsor of this session was the Northern Ireland trade group. I left that alone, though I was tempted to toss a "batteries and detonators" joke out there. What's next, "Steve - sponsored by the Afghan Warlord development counsel"?
The user count looked lower to me - after starting to draw really good crowds in Phoenix the last few years. I didn't see the numbers (I don't believe them anyway, I think they let IDC come up with them!) but it sure looked like I knew everyone there for the most part. It will be interesting to see the end tally. San Diego is tough to get too for a lot of the U.S.
Other ESG'ers have reported the show has been positive overall.
The most interesting thing I saw yesterday was a 20 minute demo in the press room from Trusted Data, Geoff Barrall's new gig. Geoff founded BlueArc, and has apparently been tainted by the big enterprise play, so this time he's after the SOHO.
Anyhow, they have come up with sort of the perfect RAID implementation - it works with any size drives, mixed and matched, automatically distributes data to protect itself - all in the background, and can be upgraded real-time. If you have a bunch of 400GB drives and need more space, plop in a 500GB disk - the system just deals with it. Also cool is that you can use the new space immediately, it does all it's re-distribution in the background, and it does it really fast because it only moves true used blocks, not every block as all other implementations that I know of do. That means a re-build will take the same amount of time if the drive is 100% full, but if it isn't, the percentage moves down to whatever level the utilization is. If it's half full, it takes half the time, etc.
These guys are hoping to make storage so simple that it will end up in our Tivo's and home media centers. I couldn't find a reason why they should stop there - why wouldn't big array vendors want this kind of functionality in their arrays as well?
Had a few cocktails with Flavio Santoni of LSI. Not sure but I think he could be the nicest guy in the business.
Dave Hitz of Netapp fame was nice enough to buy me dinner. Frighteningly smart dude. It took me several glasses of wine before I was able to expand my mind far enough to keep up with him. He may not feel the same. The most intriguing issue discussed - has Netapp become the "establishment" that it once turned on its head? And is it possible to become a success and not evolve into a modern example of what it once was that you tried to destroy?
More later.....
In my previous Webex hating blog I should have pointed out that it is not ONLY Webex I hate - but all Webex wanna-be's as well. Sorry for any confusion I may have caused by letting the competition think they were off the hook.....
|
 |
|
