Interesting Performance Data

NetApp announced some interesting SPC-1 benchmark data today. They compared their mid-level FAS 3040 with the EMC Clariion CX 3 Model 40. The results are published here:

Interesting points:

1. NetApp's configuration ran RAID-6. I'm not sure I've ever seen another storage benchmark where someone ran anything but RAID 1 or 10 because that's the highest performing. That also means that by default they get much better utilization of the space, which plays in a lower overall cost (though all that's listed are list prices, and lord knows you have to be from Pluto to pay list price).
2. Both systems ran roughly equivalent natively – though Netapp did outperform the Clariion 31,000 to 25,000 IOPS. Neither is lighting the world on fire, but with fewer disks running RAID-6 they were able to beat a pure block device – in a pure block test. I've heard them say it before, but I guess I always had my doubts that it was possible to do that running through a file system.
3. If you use the systems the way intended – with Snapshots, Netapp squished the Clariion. On their own box they created a new snap every 15 minutes, keeping a rolling (3) and deleting the old as they went. Their performance stayed pretty much the same. The Clariion created only 2 snapshots during the 3 hour test, and deleted 1 – and lost over 60% of its performance – going from 25,000 to 9,000 IOPS. Yikes.

Two take aways:

A: Netapp appears to have legit block performance, and shouldn't be dismissed because people (like me) presume it can't be true. It looks like its true. Better yet, they don't lose anything by creating, using, and deleting snapshots, one of the most important functions any storage manager has at their disposal.

B: EMC has some work to do on their own snapshots.

 

Disclaimers: I don't know the configuration and use case and fully expect EMC to pooh-pooh it in some way. I expect you'll see commentary or questions as to how full the NetApp filer was during the test, as EMC will tell everyone that if the file system is 80% full then the performance of the NetApp box will stink. I don't know if that's true, but it is what they will say. I welcome the logic and argument.

There is one guy who is the only authorized "independent" auditor of these tests, and I have no idea who he is or what his motivation is, other than the monopoly he appears to have created for himself. I'm jealous. I don't care so much about the overall numbers, what I do care about is finding out that Netapp can really run block services – not only without sucking pond water but beating a pretty darn good machine at its own game. I always knew the Netapp snapshot was a killer feature, but also sort of assumed it would take more of a toll than it apparently does.

EMC doesn't participate in performance benchmarking – and I can't say that I blame them. Someone can always build a better mousetrap designed to win a benchmark. It's chasing ghosts. Anytime a vendor can do their own benchmarking, people will be suspect. Having said that, I don't think Netapp cares about "beating" the CX – I'm sure they are happy just to show that popular opinion (or at least mine) is completely wrong – they clearly have legitimate block performance. They want a seat at the table, and this clearly shows they deserve that shot. The fact that they are demonstrating that their snap technology is 99% efficient and the Clariion's so inefficient is the most interesting thing to come out of this (I'll happily let EMC show me how I'm interpreting this incorrectly). It was my friend Chuck Hollis of EMC who blogged on the silliness of SPC and other benchmarks back in August, 2007, who said "We've never done an SPC test, and probably will never do one.  Anyone is free, however, to download the SPC code, lash it up to their CLARiiON, and have at it." I happen to agree with the overall assertion of the blog, but don't think Chuck could have figured those tricky Netapp folks would come at it from this angle. If performance isn't really an issue, then the focus moves to functionality, and snap is one of the most used (for good reason) functions in the storage world.

Finally, my own guys are going to stick their noses into this one. Tony Palmer, an ESG Lab guy and former EMCer, has already been telling me that the Clariion snapshot needs to be tuned to get good performance, but my point is that if that is true, why not make it known? This was a customer installable configuration I assume.

So let's go to work people. I need a little pre-superbowl excitement.

Value

I'm having trouble figuring out just why anyone would ever want to provide Wall St. quarterly guidance. I don't see any upside in it. I think folks will be much better off just providing guidance for the year and never saying anything other than "we believe our annual targets are accurate". Every CEO should take Bill Belichick conference call lessons.

The street is going to punish you for hitting your number, for missing your number, for saying a number, or even implying a number, so why bother? You can't control valuations, you can only control what model, or bucket, you are jammed into. Once you are in that bucket, you can never get out, so chose carefully. Why is LSI still valued as a chip only company? Why are chip companies valued lower than donut makers? Who knows? Times change. All I know is that it doesn't matter if a company is well run or not, if the market is growing or shrinking, or if the competition is winning or losing – valuations don't have anything to do with that. Valuations of companies with 20 year histories of success can be crushed in 90 days. That's just dumb.

IBM paid $330 million or so for XIV – who makes scale-out clustered block storage. They have $11 in revenue and 40 customers. IBM is betting that the technology is what the world needs in the next era, and that in their hands they can exploit the opportunity. Isilon has a zillion customers, is the originator of scale-out NAS systems, is running at $100 million in revenue, has to be close to 10,000 nodes shipped, and has $85 million in cash, but according to Wall St. they are only worth $322 million. 9 months ago they were worth a trillion. I don't get it.

VMware did $412 million last year, beat earnings expectations, grew 80% and lost 23% of its value today (only $10 billion dollars, it's not like it was a big number or anything). Why? Apparently the street expected top line revenues – one of the dumbest things you can measure in the real world – of $417 million. So for missing the top line by 1% they got slaughtered, even though they made more money – and beat the bottom line. I give up.

Here's my advice – screw 'em. Run your business. Make money. Operating with the intention of making Wall St. happy is a losing proposition. It's like being married to Sybil – she might be happy now but can easily stab you in the throat an hour later. I know it's easier said than done, especially with the pressure of all the VC's desperate to regain some semblance of a portfolio after all the ridiculous mistakes they made over the last 7 years, but hold them off.

The model for success is:

1. Say nothing.
2. If you feel compelled to say anything, only talk about how you play in the next next thing. You aren't a storage company, you are the dynamic receptacle for web 2.0. You aren't a server company; you are the fluid processing elements that scale to the demands of web 2.0. You aren't a networking company; you are providing the communications interconnect for the social networking infrastructure of the web 2.0 world. Then shut up.
3. When you feel as though you should answer a question, think to yourself – what would Bill say?

Q: Bill, can you explain why there are 2 open parking spots in your customer support center?

A: The numbers speak for themselves.

Q: Bill, you have grown revenue, market share, and margins over the last 87 quarters, how long can we expect this continue?

A: Our competition is extremely well managed. They come to play every day. They are tough. Guys like Tucci, Warmenhoven, Chambers, they are good. They dominate. Chuck, the manager at the Waltham Domino's could be the best we've gone up against.

Q: Bill, what are you going to do with your excess cash?

A: We are very pleased with the numbers, and continue to try to put the best people in the field who give us the best opportunity to win.

Q: Bill, how much did IBM's foray into your market slow sales this quarter?

A: The best you can do is to prepare for every possible contingency and make sure your people stack up as well as can be. Hopefully we continue to make big plays when it counts, especially in the fourth quarter. They are a good team, and I expect a good game. Hopefully we continue to do the things we need to do to come out on top.

Q: But Bill, you just announced record earnings, record revenue, and a record contract. You must be happy moving into the New Year.

A: We are just trying to do the best we can with the people we have. One or two breaks and the Domino's might have slowed us down.

Reyes sentenced….

I know, I know – but I just have trouble leaving this one alone. Last week the Brocade/Greg Reyes options backdating scandal came to an end (not really) and life in the world of business is now safe again (not really), right? Nope. Reyes got 21 months and fined $15 million clams. He remains free pending appeal.

So, knowing I should just shut up and move on, I can't. I was sure the judge would be forced to grant a new trial considering the primary witness to the primary prosecution claims of "he did it alone, deceiving everyone ever, especially those buffoon people in the finance department who were especially fooled by the scandalous, evil man…" (I made that up, but it's pretty much an accurate summation). The prosecution's sole witness within said finance department recanted her testimony, told the judge that she was pressured by the prosecution, and that she took back everything. The judge said he didn't care, that the testimony had no bearing on the jury's finding.

Here's the reason for my continuing my push against the tide – apparently the judge raised the "Enron" comparison when handing out the sentence. Enron? Are you kidding me? Comparing the flagrant theft of billions of dollars in a total global fraud on hundreds of thousands of unwitting, everyday people (stockholders and employees) for pure personal gain with a case like this is like comparing me to Tom Brady – sure we are both abnormally attractive, but that's where it ends.

In the prosecution's closing argument in the case (I read the transcript) that Reyes purposely deceived the finance department – a point hammered home 30 times in that one closing argument – making it impossible for Brocade's finance department to possibly account for those options properly. How could they account for them if they didn't know what was happening? They specifically said CFO Mike Byrd was a patsy and was totally left in the dark (which as anyone above the age of 8 knows is completely ridiculous) and had zero knowledge of such atrocities. Reyes, they said, was a mad financial genius with global malicious intent, in other words. At his sentencing, they again supposedly claimed that Brocade's option pricing practices were hidden from the finance department, and Byrd specifically, and thus they were deceived by Reyes.

Two days later, on January 18^th, 2008, the SEC filed a motion of opposition (Case 3:06-cv-04435-CRB Document 375 Filed 01/18/2008 ) against Byrd's motion to dismiss his case (which coincidently, was launched by the SEC only days after Reyes was convicted – in no small part because of the prosecution's countless references to the fact that Reyes acted alone and Byrd was totally duped), which in short claims (among other things):

Byrd's fraudulent conduct continued until he left in 2003.

Byrd's fraudulent conduct began in 1999.

Byrd's conduct helped conceal the backdating scheme.

Byrd instructed others to conceal the fraud.

Byrd's fraud helped conceal the fact that employees received in the money options that weren't properly accounted for.

So, if the case is based on the fact that Reyes acted alone, then that doesn't fit. If it were that he deceived everyone including Byrd, then that doesn't work. If it was because the prosecution's only witness claiming he deceived everyone in the finance department was deceived – and that it began in 2000, then that doesn't work (regardless of the fact that she is the one who then came forward after the fact to recant her testimony).

If you are really bored some day, you should read both of these cases, and I assure you that no matter your political persuasion, religious beliefs, your athletic prowess or your relative attractiveness, you will be mortified by the way this has all gone down. If it were a John Grisham novel, it would be too absurd to believe. This isn't supposed to happen in the good old U.S.A. What's next, Bill Gates has too much dough so we're going to make stuff up to take him down a peg or two? Doesn't that only happen in Russia?

The Battle of Words

The hardest problem for any new or would be analyst to get a grip on is always the dreaded "misquote". The fact is, a good analyst will spend an inordinate amount of time with the media trying to help them frame real issues and as such help them understand what it is they are really talking about. With this comment I refer more so to those analysts who have yet to lose the will to live, and as such still consider themselves not to be crusty old men – often resembling Dick Cheney. Most writers are briefed by a vendor, try to shower off the B.S. and then try to hit their deadlines.

A good journalist actually cares about things that are very, very hard to care about. Writing about IT infrastructure has got to be hard to ever care about – and I think that's why we turn so many of them eventually. The peace corp.? Flower shops? Wax manufacturing? Sure, those are easy. IT is so boring just talking about it will turn the Dali Llama violent. Most tech writers seem to be good people, or those on work release perhaps, and most seem to enjoy the sport of fishing thru the crap trying to find a semblance of reality. At the end of the story, however, they all need the sound bite.

So this week I had the chance to cause either a love fest or a lynch mob, with the same two companies, the same paper, and even the same journalist. I'm not sure how it all went down.

I got a call from IBM asking if I would speak to the Wall St. Journal regarding XIV. Surprisingly, I said yes. An hour later I got a call from EMC asking me if I would speak to the Journal about their new Symmetrix announcement. I said yes again. It turned out, it was the same reporter, so how very convenient, I thought. They ended up being two very different interviews.

During my annual practice manager meeting at Pinz, Milford, MA's mega-bowling meeting facility, I got the first call regarding XIV. That conversation was all about the reporter attempting to understand or quantify the outrageous rumors surrounding the parting of ways between Moshe Yanai and EMC – folks who made each other very large piles of money. There were excellent questions about broken chairs and virtual fist fights – none of which I could comment on (only because I wasn't there, and the parties that were there weren't talking). At the end of the long conversation, I felt I had adequately explained why the XIV stuff (Moshe's new thing that IBM bought) was theoretically cool, but not really a threat to neither EMC nor IBM's core high-end business – that this represented a commercial application of the "Google" storage phenomenon. He bought it, I bought it, and onwards we go. In the paper the next day, my pithy (and completely brilliant) comments were omitted from the article, and instead left to the non-pithy, somewhat less than completely brilliant commentary of others. (Perhaps I'm jaded; maybe they were both pithy and brilliant, but I doubt either referred to the Mossad). I was shunned in the editorial process. I felt ok, though, as I know I taught the young man a thing or two that he had absolutely no interest in learning.

The next morning my phone rang from the very same gentleman reporter guy. This time, he wanted to talk about EMC's new magical announcement on the Symmetrix 8 million, or whatever the current model is. This time it was about a technology component – namely flash inside the uber array. I spent the same hour attempting to pass on decades of wisdom that no one cares about (including my wife and kids, dog, cat, and mailperson) to find a point he could rally around. I came up with this: "in these circles, performance matters. There are a certain small number of folks who simply can never get enough of it, and are willing to pay thru the nose to get it. If, in those application environments, the performance delta is as I suspect – large – then it means that IBM and HDS will be behind the eight-ball until they respond in kind". Simple, factual, and accurate. They printed "HDS and IBM are screwed". (Subtle paraphrasing on my part). Now, in certain places, IF the stuff works, and IF folks buy it, IBM and HDS could be screwed for a while – until they address the issue. The problem is no one is screwed forever, only until the next marketing cycle. The battle of the word is a hard one. If Andy Monshaw cried on TV, would people go back to Shark? Hey, it worked for Hillary. Plus, Andy is more believable.

So go easy on your friends who try to tell the truth. Sometimes an editor who doesn't know Flash disk from Flash in the Pan (walking, walking, in the rain…..) is cutting out words to fit the page. Stop shooting the poor teacher. The day before I spend a ridiculous amount of time attempting to educate the reporter on the fact that there might be some merit to IBM buying XIV (instead of for spite and other ill advised emotional outbursts), IBM folks were none too happy with me for being misquoted on the XIV technology saying it could eventually replace the DS8000. It could do that, but my quote was that this technology, once proven out, should be leveraged across the back end of every storage product IBM has, DS included. They probably would have been happy had that been printed.

So, in conclusion, don't put all that much stock in what you read, as the process is designed to be incomplete and often downright inaccurate. We all do the best we can……… (He said what about his "can"?) Go Pats. (He said he wanted to pat his can?).

A Sign of the End?

IBM buys XIV and the next day Netapp buys Onaro. Israeli storage startups, once as abundant as Canadian snowflakes, or goose droppings on a nice golf course, are becoming increasingly rare. We tracked 24 of them in 2001. Now we're down to only a handful. Exanet and Files-X are the only old-timers left I think. Continuity, Diligent and StorWize are relative newcomers still in the fight. Sepaton was Sangate 45 years ago, but now is about as American as Brittany Spears so they don't count. Diligent will be the next to go – with the run on bulk storage plays designed to support the new era of digital content, a big honking de-dupe data protection play for big data centers like them will get scooped up this year. I don't see how Continuity lasts for more than 1-2 more years without getting bought either, and if the StorWize stuff works as advertised then they just have to wait. Folks that help other folks stuff 800lbs. of junk into a 2lb. container are finding themselves more and more popular. Exanet can make a claim in the envious world of mega-scale, and will probably spin themselves up as the file equivalent of what XIV brought to the table. I'm not sure exactly what Files-X is up to these days (but still hold hope for them as they were founded by my pal Jacob Herbst and we are bonded together by 9/11). Where will we be then? Maybe we should start a pool as to who will be the last standing. "The last of the Israeli Storage company's" sounds like a boring tale but it worked for "The Last Emperor" and "Last of the Mohicans".

Can it really be that there are only 5 active Israeli storage players left? Surely I'm missing someone, but when in the last 20 years has there been such a dearth of Israeli companies in the space? When XIV's Moshe Yanai became an international sensation after creating and riding the wave that was Symmetrix at EMC, it seems every former Israeli commando turned engineer was starting or joining a storage company. There has been more money per person generated out of the storage business within Israel over the last 20 years than any other industry. (Yes, I completely fabricated that fact but I bet I'm darn close).

Israel has one of the best educated populations on the planet. They seem particularly adept at engineering. Love or hate them, you have to acknowledge that as a gross overgeneralization – they are smart. If one were to observe the rush to the storage industry by the Israeli's 20 plus years ago as a harbinger of what was to come, and placed some bets in that area, one would be filthy stinking rich right now. Should we assume that since the numbers say the Israeli's are no longer interested in the Storage industry, that this is a new harbinger we should consider?

Don't get me wrong, it could be coincidence, or the fact that from 2000-2006 the stock market became so anemic that startup funding in general hit an all time low and we're just seeing the result of that, but I'm not sure. It could be that we truly are a global industry today, as exhibited by the fact that there are lots of U.K. and other European technology efforts under way. Heck, China is starting to innovate in the area.

I hope I'm just paranoid, as I do enjoy being briefed on technology by people who have the ability to kill me a hundred ways with a napkin.

IBM Buys Moshe and the Sunshine Band (XIV)

IBM announced this morning that it acquired Israeli storage company XIV, founded by Moshe Yanai back in 2002ish. Moshe is the father of the Symmetrix, designed to destroy IBM's then stranglehold on the high-end disk market – which he was pretty darned successful at doing. Along the way, one could argue that that was the birth of the entire "block storage as a separate market" as well, making a lot of people a lot of money and giving me something to do in life.

I was literally having a brief we wrote on XIV posted this morning, so you can get our take there.

As far as the IBM thing goes, it will be interesting to see. IBM isn't immune to the fact that there is a whole wide world of brand new digital content being created every day and they are going to have to participate with new products and technologies in order to play. The good news is that IBM is not playing catch up this time, as no one is really in the lead on this new set of "Infrastructure 2.0" opportunities, so with their brand muscle they have as good a shot as anyone to make the rules. XIV in the hands of a global IT behemoth like IBM could be a powerful play – or they could screw it up – but at least they get the opportunity to dance this time.

They won't be alone for long, but any head start is a good head start if they have a plan to push forward quickly. Finally, it seems to me that this play has the opportunity to extend way beyond just another "storage" buy. This is an architecture that fits into IBM's data center of the future pitch seamlessly, carries huge services potential (assessments, data migrations, and ongoing data lifecycle management and protection), and has the ability to wrap a ton of other IBM product/technology around it. If IBM can use a this architecture as a reason to have the "why are you doing that?" conversation with high-end IT professionals (around treating fixed digital content the same with they treat transactional information) and get it to "we can show you how to be much more efficient and effective" then part of the ultimate solution will end up opening up opportunities for everything from FileNet to Tivoli.

More to come……