1 2 3 4 13 Previous Next

Dave's Blog

192 Posts

Are Corporate Values Yucky?

Posted by hitz Apr 25, 2011

I gave a TEDx talk on the question of whether corporate values are yucky. My initial answers were: (1) Yes, (2) Sometimes, (3) I used to think so, and then I elaborated for fifteen more minutes.




Culture of Innovation

Posted by hitz Mar 28, 2011

What does it take to create a culture of innovation? That's what I discuss with Criss Marshall in this edition of iNside NetApp.


Today NetApp grew past ten-thousand employees.


Of course, there’s no particular magic in that number: just a coincidence of base ten math. Still, I’m always interested in crossing boundaries, natural or artificial, and I do think it’s important to think about what size means.


In particular, many people appreciate NetApp’s culture, and they ask, “How can we make sure our culture doesn’t change as we grow?”


I -- on the other hand –- believe that our culture must change. When a company grows, what is appropriate changes. For instance, in a 10 person company, you don’t need formal all-hands meetings with follow-up video-on-demand to communicate with everyone. That would seem stiff and unnatural. You just invite them all to lunch. But as a company grows, it becomes much more difficult to share information, and it is important to create more formal employee communications. We have internal web sites with notes and videos from executives, as well as formal notes to help managers communicate to their teams. In a small company, that would have felt like overkill –- also overly formal -– but now it is appropriate and important.


I have heard culture defined as “values plus behavior”. Values are the part that shouldn’t change, or at least not too often. Culture, on the other hand, should change. It should change in response to the environment around us –- in response to economic conditions or competitive conditions, and it should also change as NetApp itself changes.


Even if ten thousand employees is an artificial boundary, it seemed like a good time to remind ourselves: As NetApp changes, it is important that our culture changes. We still want to be the best company we can be, but what is best at 10,000 is different than what was best a few years ago at 5,000. What stays the same is the core values that the culture is built on.

NetApp has just agreed to acquire Engenio, which is the external storage systems business of LSI Corporation. An obvious question is how this will affect our unified architecture. Will we now suffer from the disadvantages that EMC faces with its fragmented product line? We don’t think so and here’s why.


The problems of multiple architectures stem from having different features in different product lines. How many different cloning mechanisms must a customer learn? How many times must the engineers develop deduplication? How many versions of dedupe must they maintain? And of course, the same questions apply to thin provisioning, compression, Oracle integration, Exchange integration, and on and on and on.


Here’s the key point: this type of advanced data management has not been the primary focus for Engenio. Their RAID array is enterprise quality, highly available, very dense, very high bandwidth – and also very simple.  This makes all the difference. Let’s examine some pain-points of having multiple architectures in more detail, and consider whether they apply to NetApp with Engenio.


(1)  Hard for Customers to Learn

The customer must learn how to deploy and manage multiple implementations of the same advanced data management features. Not with NetApp. ONTAP has the features and Engenio does not.


(2)  Hard for Vendors to Develop

The vendor must develop and maintain multiple implementations of the same advanced feature. Not with NetApp. We add them to ONTAP and not to Engenio.


(3)  Hard for Customers to Choose

How do you choose which product to buy when each architecture has different features? This is especially confusing for a shared storage infrastructure intended for use with many different applications. There can be painful tradeoffs if one product – say – supports dedupe for NAS but a different one supports thin provisioning for SAN. Not an issue for NetApp. If you want the advanced data management features, go with ONTAP which has them all.


I’m over-simplifying. There will surely be some overlap, but NetApp and Engenio have very different design points optimized for different workloads and market segments. Having one advanced storage architecture and one basic RAID array is very different from having a portfolio of advanced storage architectures, each with different features, some in common and some not. (Finally, whether or not you believe my analysis, “two” is much simpler than “lots.”)


So why did we buy Engenio? What is the opportunity, and how do we expect it to evolve over time?


The observation is that, while many customers and workloads do require advanced data management, some need “big bandwidth” without the fancy features. For them, the best solution is a very fast RAID array with great price/performance. Perfect for Engenio! Two immediate opportunities are Full Motion Video (FMV) and Digital Video Surveillance (DVS), and over time we believe there will be more. Engenio didn’t have the partners or the sales force to reach some of these markets, but NetApp does. And of course, Engenio is also a successful business with a great team of talented folks and an install base of 350,000 units. So much the better!


For me personally, there is a much more interesting long term question: What storage arrays should we use if we move ONTAP into a virtual machine? I expect this question will surprise some people. What!? Isn’t NetApp a hardware company? Surely they wouldn’t even consider selling ONTAP as a VM! Not only would we consider it, but we are doing it. The first version is a product developed in partnership with Fujitsu and it is shipping now.


I don’t know the future of ONTAP in a VM (aka Virtual Storage Array or VSA), but let me share a story. Many years ago, several startups developed storage controllers that went in front of other people’s storage. They argued, to their VCs, that there would be no competition from us “because NetApp would never ship storage controllers without disks.” We surprised them by creating the V-Series, which lets you run ONTAP on top of storage from many other vendors, including EMC, HP, IBM, Fujitsu, Hitachi, and – of course – Engenio. The V-Series started small but eventually became a major product line for us. If history is any guide, running ONTAP as a virtual machine could become a big business and also a great match for Engenio storage. This isn’t an announcement – I’m just trying to give a general sense of the possibilities here.


PS: Last week I (finally!) got around to signing up for twitter. @DaveHitz.





This is a continuation of the  discussion with ESG senior analyst, Mark Peters about what’s going on in IT.   The first two posts in the discussion series can be found here and here.

Dave Hitz:    When you  talk about consolidation and standardization, it seems to me the way IT is  playing out is that whatever resources you have, or whatever technologies that  you have, you want to be able to share them across multiple users.

That  need to share drives the need for other things, as well.  If you’re going to be  sharing, you need to be able to share securely.  You need to be able to share  reliably. 

But in a sharing context, because you’ve got different users  with external clouds, you may have different companies sharing the same  resources.  It really brings a lot of other kinds of management requirements to  the fore.

Mark Peters:  Sharing is another way  of saying being efficient.  It’s another way of saying  consolidating.

That aligns with everything I mentioned about using the  minimum resources you can use to get a given piece of work, or set of works  done, to the level that you need it done – still with the quality and all those  other things you just mentioned, security, etc.  Sharing is absolutely the way  ahead.

There are two things that are making this really crucial as you  look forward.  One is growth.  The demands on IT have grown beyond anyone’s  expectations.  At the same time we’ve had this massive economic malaise around  the world.  The combination of those two things has made everyone go, whoa, hang  on a second.  We are not running IT in a sensible fashion.  We’ve been doing it  focused almost 100 percent on being effective without thinking about being  efficient.

We can’t go back to that because the economic issues, and  perhaps the emergence of certain technologies, has enabled us to turn, to focus  on efficiency, whether that’s via long-term flexibility, sharing or some of the  other things we’ve talked about. 

Dave  Hitz:    What is the percentage of the overall economy that will  be the IT industry?

For many years, that percentage has been going up  and up and up.  People just keep spending more and more and more on IT. But you  get to a certain size, and you say, we’re just not going to double our IT spend  in our company from two percent to four percent, and then four percent to eight,  and then eight to 16.  Then what?  We’re going to double it to 32 percent of our  whole business is paying for IT?

We’ve got to stop.  It’s a little bit  like health care in the US.  You look and ask what percentage of the total  economy can be health care?

Are people, as part of their re-evaluation,  saying, look, IT is not going to grow anymore.  In fact, maybe it should shrink  a bit.

Mark Peters:    We’ve got to do more  with less. What’s interesting about that is we can absolutely do that, even  today.  I’m sure it’s only going to get better.  If you look in the big trends  over the last few decades, we haven’t really worried about IT expense, because  it’s produced so much more value for businesses than it was costing, it was easy  to justify spending money there.

Whether it’s the economy or the  unsustainable growth percentage of the companies that you discussed, whatever is  causing this, we’ve realized it’s not that there are lower returns from IT, but  like any other part of the business, it’s got to justify those, and it has to be  producing results for the business.

IT has to do more – and whether it’s  for less or the same, or whatever – but it has to do it in a more efficient  manner.  The whole business, the whole nature, the whole architecture of IT – if  you take a really big view of it over decades – has grown up in a very  inefficient manner that got the job done really well.  And now we just need to  tighten it on both sides of that equation.

Dave  Hitz:    We’ve been talking a lot about the importance of  flexibility and the importance of being ready for what comes with the future.   What is the business benefit of this in the end?

Is it really all about  saving money?  Or are there other things that come out of this  approach?

Mark Peters:    You’re right to raise  that question.  It’s a great deal about saving money, but you remember early on  when I said future flexibility, or being ready for the future is not about IT,  it’s about the business.

The flip side to this is, if it’s done well, if  you know where you started and you know the impact and potential impact of  things you either do or might do, then what that drives to is producing money  for the business.

Most people will cite that as faster deployment of  applications.  Yes, that’s great.  But you need to go one stage beyond that,  because that’s IT centric.

What did the faster deployment of that  application buy the business in terms of bottom line - like competitive  advantage or customer satisfaction, things that can be measured in terms of  revenue?  That’s the flip side to this equation.

Dave  Hitz:    So, what I hear you say is to do not just a good job, but  a really good job at IT, you need to understand what an application means from a  business perspective.  What does it do for customers?  What does it do for  revenue?  What does it do for profitability?  If you as an IT person can’t make  that linkage, you’re not operating at your full potential.

Thank you for  your time Mark.

Mark Peters:    Thank you.  It  was a great conversation.

This is a continuation of the discussion with ESG senior analyst, Mark Peters about what’s going on in IT.  The first post in the discussion series can be found here.

Dave Hitz:    In the quest to become Future Ready, what are the key issues that an IT organization ought to pay attention to?

Mark Peters:    The thinking and planning that I mentioned before are 1A and 1B on the list.

You have to know where you came from and where you want to go, to the best of your ability, but keep flexible so you can react.  For instance, we know there are going to be more virtual servers.  We know there’s going to be more storage.  But what are those things going to be asked to do?

The second issue is, how do you quantify what you’re getting from these various technologies or approaches? It’s easy to throw out things like speed and efficiency.  But you need to stop and think about those for just a second.

There is a discussion in the marketplace about driving cost out of the business.  It has been crucial for the last few years, and continues to be so.  Now you’re going to have this debate between using multiple vendors, the best of breed approach, if you like, and more of the collapsed stack or converged approach.

The question users need to ask themselves there is, does one make me lose as much or more as it makes me gain. Effectiveness, in my view, is getting the job done as well as you possibly can.  That’s great, and we should all do that in IT, whatever the future is, whatever the flexibility is. But efficiency is using the minimum number of resources you can to get that given work done – still to the same standards, still as effective, but doing it with fewer resources.  You can only do that if you know where you’ve come from, what you were doing, what the business benefits of all those things were, and therefore be able to quantify the improvement, because otherwise you’re just doing more of the same.

The only thing installed in every data center is inertia; unless you have a clear path that shows not only where you’re going but also where you started from.

Dave Hitz:    When you start to think about technologies, what technologies do you think are going to be involved?  Let’s imagine somebody who is just beginning the planning cycle and they’re saying, our current data center is full.  We know we’re going to have a new data center.  And we have an opportunity to think afresh about exactly what we’re going to do as we plan where we’re headed.

Mark Peters:    We talked about being future-ready, and you talked about knowing certain things that were going to happen, but we don’t know exactly how they’re going to play out.

If you’re talking about right now and the attributes that are important, you’ve got to look for things that have an ability to consolidate, an ability to use virtualization to the fullest extent possible and an ability to use standards wherever possible.

Standards are about ease.  Virtualization is about efficiency, which is what I just said is so important.  Consolidation is just about pure economics.  And overall, you’ve got to be able to manage this.

Because you’re making some decisions in the next few months, you’re going to want more of a unified platform.  That is not just a wave in the industry, but it’s a very logical approach.

You’re going to want to talk with people who are utilizing solid state and know how to do it, because, at the end of the day, solid state is all about serving (IO); in other words, getting back to that efficiency of minimum resources for given work.

The only reason you store things is because, ultimately, you may need them.  And therefore, the choice is, how frequently, how often and when am I likely to need it?  And that will determine where it goes.

If it’s about getting immediate work done, it should go on solid state.  And even though that’s a very bold statement, that’s true as you look for the medium future.  Right now we have data on different spinning disks as well because that’s the technology we have available.

The cloud is about how you consume IT and how you consume resources.  And whether that’s done behind your firewall or from a public source, it again comes back to this flexible use of resources; it comes back to consolidation, standardization, and so on.  It may not be suitable for all.  But it’s another way that, whether it’s internal or external, you can drive efficiency and leave yourself flexible.  Those are two things that are very important.


Stay tuned for Part 3 in the discussion series, which will cover the goals of future-ready.

Dave Hitz: I’m Dave Hitz and I’m here with ESG senior analyst, Mark Peters.  Over the course of the next three posts we’ll be talking about what’s going on in IT. What does it take for an IT organization to future-proof their environment?


Mark, good IT people are always worried about what they need to do to future-proof their environment.  That’s just part and parcel of what you do if you’re an IT person.


In the past few years it’s become especially important, given how trends are playing out.


Do you see things that way?


Mark Peters:   Yes, I do, Dave. 
I think it’s about planning.  That may sound too simple.  But we’re all good at thinking about growth in terms of capacity.  We’re good at thinking about the speed of our devices improving. 


But what we have to think about when we’re planning – and this gets to the future-ready aspect that you’re talking about – is about being able to change, being able to change fast, changing on the fly as business changes.


Dave, do you remember when IM came out, instant messaging?


Dave Hitz:    Yes.


Mark Peters:    I know that the Yahoo! IM team planned their maintenance between nine a.m. and five p.m., because, that was when the system wouldn’t be used much!  Now, in many businesses, they actually run on IM.  Things can change very fast.


The second point about this planning is to be future-ready.  It’s not just about the future happening to you as it evolves.  It’s about making it happen in the way that you want it to. 


In order to ‘do’ future-ready, you need to do ‘past-ready’ too, which means knowing where the heck you are now, understanding everything from utilization to power, to how many dedicated systems you’ve got.  How well are they used?  What are they producing for the business?


These are questions that, because they’re things that you’re already doing, you can answer.  And then you can talk about planning for the future.


Dave Hitz:    One of the things that’s so hard about planning for the future is you don’t know what’s going to happen.  Even though we don’t know exactly what’s going to happen, there are a few statements you can make that are straight-forward.  For instance, there are going to be a lot more virtual servers out there.


It’s not clear how many of them will be VMware and how many of them will be (hyper-V) or Xen.  There are dynamics in there that are hard to predict exactly.  But it isn’t hard to predict there will be a lot more virtualization, a lot more apps will be running in virtualized environments.


If you take that as a starting point you can do quite a bit of planning, even if you don’t know anything else about how it will play out.


Mark Peters:    Yes, I completely agree.  What you’re really talking about is the essence of what we know will happen.  But there’s the second layer, subtleties and specifics that we don’t know.


Therefore, what you need is something that is going to be flexible as you go forward.  You know you need more virtual servers.  You know pretty well how to deal with that.  But somehow within this, you’ve got to keep some flexibility.


And what’s interesting about the whole notion of flexibility, when you’re talking about IT is, in many respects – and this goes exactly to what you were just saying – it’s not about IT itself.  It’s about business.  It’s about opportunities.  It’s about applications.
Dave Hitz:    If you go back to the early ‘80s, when the PC was first introduced, people were saying, the PC is going to change everything, and it’s going to have a radical impact on business.  And that was true.  No question about it.

People were also saying, this will lead to the paperless office.  I haven’t noticed paperless offices in my organization.  How’s yours doing?


Mark Peters:    Exactly.  Although, the nature of the paper changes, doesn’t it?  We all print things off.  If you’re old enough like I am, sometimes you need it on a piece of paper to take it in, because you still don’t take it from the screen properly.

But, no, we haven’t gone paperless at all.


Dave Hitz:    The nature of the paper has changed.  That is what I find so interesting.  We were completely wrong about so many details.  And yet, big picture, we knew what was going to happen, and it did.


When people predict how cloud computing will play out, and how virtualization will play out, I suspect that our guesses – yours and mine as industry watchers – are about as accurate as the paperless office, which is to say, completely wrong and completely right, at the same time.


Mark Peters:    Absolutely.


There is a very interesting aspect to this I think we, as IT practitioners, need to think about.  The way we organize people and the way they’re rewarded and paid.  It’s one of those things I don’t think gets enough attention.


Now, you and I can drone on all day about flexibility, and that’s absolutely vital.  But there has to be flexibility within the way that people are measured and rewarded as well.  Because the lack of that is one of the reasons we’ve grown up with so many distributed, dedicated systems.


Everyone likes world peace.  Everyone will check the box for that, but no one wants to go and fix it.  It’s exactly the same in IT.  Everyone will check the box for ‘let’s share everything and let’s integrate everything and consolidate everything’. Oh, yes, well, except for the stuff I do.  I’d like my own server, and my own storage and my own application, thank you very much, because that’s the way I get paid, because I get measured on whether Exchange is up that day, or whatever else it might be.


It’s very simple to start a conversation about the future and flexibility.  But you really need to stop and think about what that means and how you do something about it.



Stay tuned for Part 2 in the discussion series, which will cover the key issues in the quest to be future-ready.


iNside NetApp with Dave Hitz

Posted by hitz Sep 14, 2010



Always Bet AGAINST Your Employees

Posted by hitz Nov 28, 2009

Recently a group of NetApp managers tried to motivate theiremployees by betting that the team could meet a particular project milestone. Ifthe team missed the date, the managers promised to dye their hair brightcolors. (Silly hair bets have a long history at NetApp.)


Unfortunately, the team missed the date, and as a result,the managers not only had to deal with bad news, but they had brightly coloredhair to advertise it. When people would ask, “What happened to your hair,” themanagers had to say, “I bet my team that they would hit their schedule, butthey missed.”


Even if there are perfectly good reasons for the miss, thelistener might still conclude: You notonly look like a clown, but you manage a whole team of clowns.


Much better to bet that your team will fail. When I firststarted running engineering (back in 1999), product reliability was an issue,so I bet my employees that they could notimprove quality by a factor of 10. If they did, I would dye my hair any colorthat they wanted.


Imagine how this played out. The team succeeded (as measuredby a man who cared so deeply about quality that I knew he would never cheat),and I ended up with hair that was bright magenta, blue, and red. To gethair these colors you can't put it on top of brown; you have to start bybleaching to white. So I looked like a freak, but at least, whenpeople asked me about it, I had a good answer: “Let me tell you about myawesome team. I should never have bet against them.”


The conclusion: Youlook like a clown, but at least you manage a good team.


So I say, always bet againstyour employees. Of course, it helps if they understand it’s a bet you want to lose.


Why NetApp Is Winning

Posted by hitz Nov 23, 2009

NetApp recently released financial results, and according toone analyst, “Itwas a supersonic quarter.” Never mind the financial details – I think themore interesting question is: What’sgoing on? Why is NetApp Winning? (Those who like details can check here.The quick summary is that our revenue was flat from a year ago, up 9% from theprevious quarter. Profitability has also recovered.)


Two interesting trends are combining to boost our growth. (1)Many companies have an aging IT infrastructure and can no longer postponecapital purchases. (2) Server virtualization forces companies to consider newIT architectures, and that plays well to NetApp’s strengths.


When the dot-com crash hit, back in 2000/2001, manycompanies had recently updated their IT infrastructure as part of Y2Kremediation projects. Internet companies and their suppliers had been growingso quickly that most of their equipment was new, and many of them went out ofbusiness so the market was flooded with “almost new” gear. It was a toughrecovery for IT vendors. This time is different. This downturn began gradually,perhaps in 2007, so people have already been postponing capital purchases forquite some time. Partly this is intuition, based on talking with customers, butone metric we can track is one-year service renewals. In good times, peopleupgrade their equipment when their initial service contract expires. In badtimes, they want to defer capital spending, so they simply extend the servicecontract by a year. In the past year, we have seen many one-year extensions,and we believe this reflects lots of pent up demand.


So the first key trend is that IT infrastructure is aging,and eventually companies will need to buy.


The second key trend is that server virtualization isforcing people to consider radically different architectures. I’ve had many customerstell me that their initial goal was to install VMware, but they discovered thatin order to get the full benefit, they first needed to re-architect theirstorage infrastructure. NetApp was an early leader in storage for virtualizedenvironments, so in many cases people have switched to our storage even whenthey were completely satisfied with their previous vendor. It’s not that theold vendor did anything wrong; it’s just that their products didn’t work sowell with server virtualization. Virtual servers can be created quickly, so youneed storage technologies like thin provisioning and cloning that let youprovision storage equally fast. Virtual servers proliferate, so storageefficiency techniques like data deduplication become all the more important.The whole point is to save money, so good automation is critical. NetApp’sstrengths perfectly match the requirements of server virtualization.


It’s hard to sure what’s happening. It could simply be thatIT spend is picking up, and a rising tide floats all boats. If this is true,then you should expect similarly strong quarters from our competitors. On theother hand, it could be that NetApp’s strength in virtualized environments isallowing us to get much more than our fair share. It’ll take a few morequarters of results to be sure, but this is what we think is going on.


The Petabyte Era is Dead

Posted by hitz Nov 19, 2009

Sometimes numbers have a symbolic value beyond any realmeaning they contain. Like when your odometer rolls over to a hundred thousandmiles. Of course, it’s all arbitrary, since a

mile is just a certain number ofRoman soldier footsteps, but even so, in a funny way, it feels like a major achievement.


At NetApp, we just had one of those magic moments:in the past 52 weeks we have shipped an exabyte of storage. That’s athousand petabytes. Or 1,000,000,000,000,000,000bytes. Look at all of those zeros! It has no real meaning, but I still love it.


The petabyte era is dead. Long live the exabyte era!

I got a reader comment so perfect that it calls outto be shared:


I'm not the most technical tool inthe shed, but I read a lot, and I have been reading a lot about cloud computing– and the lyrics of a song keep playing in my head:


"I've looked at clouds from both sides now
From up and down, and still somehow
It's cloud illusions I recall
I really don't know clouds at all"


Cloud Computing – how does it affect me, the person on the receiving end (securityissues, reliability)? The definition of Cloud – Webster's dictionary:


  • Something that darkens or fills with gloom.


  • A dark region or blemish – something thatobscures.


Again, average person with concerns.


Posted by: Evelyn Lindquist | October27, 2009 at 10:41 AM


I’ve noticed that technical people sometimes lovedefinitions that are so intricately detailed that non-technical people –including most business people – can’t understand them. I think Evelyn’scomment is a sign that cloud computing is suffering from this disease.


There are good reasons for people providing cloudservices to dig into the technical details of what they are doing and how theyare doing it, but I think that we need much simpler definitions for people usingcloud services.


I just can't let Evelyn's definitions stand (gloom, dark region, blemish), so here is my attempt at a simple, non-technical definition of cloud computing:


If I am a business person and Ihave a business problem that can be solved with IT, there are two ways to goabout it. The traditional way is tochose an application, find some hardware to run it on, find some storage forit, find space and power in my data center, find people to operate it, and soon. The cloud computing way is tofind a service on the internet instead of an application I run myself, and tolet someone else handle all of those other steps. I don’t own a data center,buy any equipment, or operate anything. All my capital expenses are convertedto a service fee, or – more common in clouds for consumers – the service is freebecause I have to look at advertisements.


I’m not saying that technical people should ignore theimportant differences between various cloud approaches (Software-as-a-Service,Infrastructure-as-a-Service, Platform-as-a-Service, Storage-as-a-Service,Internal/External/Private/Public). But I do believe that we’ve got to figureout how to hide as many of these intricacies as possible from the non-technicalpeople who just don’t care.

Filter Blog

By author: By date:
By tag: