NetApp Sues Sun for ZFS Patent Infringement

This morning, NetApp filed an IP (intellectual property) lawsuit against Sun. It has two parts. The first is a “declaratory judgment”, asking the court to decide whether we infringe a set of patents that Sun claims we do. The second says that Sun infringes several of our patents with its ZFS technology.

How did we get here?

Like many large technology companies, Sun has been using its patent portfolio as a profit center. About 18 months ago, Sun’s lawyers contacted NetApp with a list of patents they say we infringe, and requested that we pay them lots of money. We responded in two ways. First, we closely examined their list of patents. Second, we identified the patents in our portfolio that we believe Sun infringes.

With respect to Sun’s patent claims, our lawsuit explains that we do not infringe, and – in fact – that they are not even valid. As a result, we don’t think we should be paying Sun millions of dollars.

On the flip side, our suit points out that Sun’s ZFS appears to infringe several of NetApp’s WAFL patents. It looks like ZFS was a conscious reimplementation of our WAFL filesystem, with little regard to intellectual property rights. Here’s what creators of ZFS have to say: “The file system that has come closest to our design principles, other than ZFS itself, is WAFL … the first commercial file system to use the copy-on-write tree of blocks approach to file system consistency.” One of the first patents I filed at NetApp describes this “copy-on-write tree of blocks” technique in detail.

We filed suit against Sun because after we pointed out the WAFL patents, their lawyers stopped getting back to us. The first part of our suit is a declaratory judgment. It’s complicated, but the basic idea is that Sun claims we infringe their patents, so we are requesting a trial to show that’s not true. In essence, a declaratory judgment calls their bluff. It allows us to force a legal conclusion, rather than leaving this threat hanging over our heads. The second part is a complaint against Sun for infringing several WAFL patents with ZFS.

As we file this case, I’m painfully aware of the bad feelings that many people have about IP lawsuits. Business people sometimes view them as the last gasp of a dying company, while some technical folks view them as evil under any circumstances. In this case, I doubt that business people will be too concerned, given our growth rate, profitability, and the fact that we are responding to Sun’s claims against us. I expect skeptical technical people to be harder to comfort.

This case is especially sensitive, because Sun has released ZFS as open source. It is admirable to contribute to open source. I have done it personally, although it was a long time ago that I was writing code, and NetApp has also contributed as a company. But it doesn’t help the open source movement to give away code that is encumbered with someone else’s patent rights. The sooner we determine the true status of ZFS, the better it will be for everyone. NetApp certainly doesn’t believe that we can somehow erase every copy of ZFS that has been downloaded. (Impossible!) This lawsuit isn’t about downloads for personal or non-commercial use; it is about what Sun is doing.

We could have a long debate about the merits of the patent system. (I expressed my own qualms here.) But like it or not, it’s the law of the land. Here’s an analogy. Suppose you own a pro football team, and you really believe that touch football would be safer and more humane than tackle. You can lobby all you want to try to change rules, but until you are successful, I recommend that your team keep wearing pads and helmets. NetApp is participating in attempts to reform the patent system, but meanwhile we will play by the rules as they exist.

In closing, let me say that the legal system is a method of resolving disputes between companies, but it does not mean we are “at war”. During this process, we will continue to support all Sun products, including ZFS filesystems that use NetApp storage. It is not in anyone’s interest – Sun’s or ours – to leave a bunch of customers in the lurch. I hope that Sun will respond in kind.

[Note: Sun has now countersued. See Jonathan’s blog post announcing that, and my response.]

Non-technical people can stop reading now.

It is important to me that technical readers not confuse NetApp with SCO, so in our lawsuit, we provided a starting point for people who want to dig deeper. This is not an exhaustive analysis of our case. We simply highlight one particular patent and one particular aspect of ZFS to help people see that this case of infringement is real.

Here’s how the ZFS designers describe filesystem consistency:

The best way to avoid file system corruption due to system panic or power loss is to keep the data on the disk self-consistent at all times, as WAFL does. To do so, the file system needs a simple way to transition from one consistent on-disk state to another without any window of time when the system could crash and leave the on-disk data in an inconsistent state.

In the ZFS paper, search for uberblock, and compare its role in filesystem consistency with the role of the root inode and file system information structure in our patent 5,819,292. Read claim 4 and its descendents, which describe our tree-of-blocks consistency technique. Claim 8 and its descendents describe efficient snapshot creation based on the tree-of-blocks. Some more useful references are here (see pages 7 and 8), here, and here.

We hope that this level of openness will help raise the bar on how people pursue intellectual property suits.

Let me insert the usual legal disclaimer: I’ve quoted our complaint, but beyond that, I won’t engage in any public comment on the technical details of our case while it is pending.

More Links

• Sun's CEO discusses the lawsuit.
• Sun Patent Team Demanded $36 Million From NetApp

Comments

[Joshua Konkle]

Looks great Dave! Your position is very well laid out. After listening to the call this morning, it seems that Sun was standing on your shoulders and forgot why they had such a good view of the answers.

It is interesting to read you have chosen Lufkin, TX (north of Houston and Southeast of Dallas, due East of Austin). IBM used that court district for the patent infringement case against Amazon, so follow NetApp and Sun would be there as well. (http://www-03.ibm.com/press/us/en/pressrelease/20481.wss)

It's great to understand the details; doesn't look like your upcoming engineering schedules will be affected by this case. Full steam ahead for Data OnTap and OnTap GX; (http://blogs.netapp.com/dave/2007/04/is_data_ontap_b.html)

JK

[joseph martins]

Hi Dave,

Is the second part of NetApp's suit retaliatory, or had NetApp planned to pursue legal action all along? This is the first I've heard of NetApp's intent to sue for WAFL infringement - in the 5-plus years since the inception of ZFS - despite the existence of NetApp patents dating back to the mid-90s. It is a fair question given the context and timing of NetApp's suit.

Even if NetApp demonstrates a legal basis, cynics such as myself might see this as a SCO-like attempt to, at the very least, hang a cloud of legal uncertainty over a competing technology in an attempt to stifle its adoption.

Certainly, a company is obligated to protect its IP. I simply found it interesting that NetApp took so long to respond.

-- joseph martins

--------------------------------------------------------------
You asked whether NetApp planned pursue legal action all along, independent of Sun’s action. I think the most accurate answer is that Sun goaded us into paying attention. In our 15 years of existence, we have only filed one other IP lawsuit, so it’s not like we have an aggressive track record, or teams of people standing by, ready to sue.

On the other hand, I won’t pretend that we would never have sued if Sun hadn’t approached us first. We focus on innovation as a company, and we do intend to defend our intellectual property.

-- Dave Hitz

[SS]

I always wondered, why has NetApp not done this all this while. Sun has been claiming the same thing on its official blogs (http://blogs.sun.com/val/date/20040909) and NetApp never took notice till now.

I believe this was longtime pending.

[Mark R. Hinkle]

In the event that NetApp wins it's case, what are your plans for asserting your patents. If I was a user of OpenSolaris and ZFS would NetApp's actions have repercussions for me? {I personally am not I am just trying to understand the potential consequences of the suite.}

-- Mark R. Hinkle

--------------------------------------------------------------
Our interest is on commercial use of ZFS. That is, we are concerned with companies who take our IP and turn it into products that they make money on. For obvious reasons, we are especially concerned about commercial use of our IP that would compete with NetApp.

We are not concerned with downloads of ZFS for personal or non-commercial use.

-- Dave Hitz

[Fazal Majid]

And to think I once actually liked NetApp as a company...

You can kiss my business good-bye. I am adding NetApp to my list of companies to shun, along with SCO and Sony. I am deleting any testimonials I ever put online for your products, and will contact your marketing department shortly to make sure they delete my reference for the S500.

-- Fazal Majid

--------------------------------------------------------------
I’m sorry you feel this way. I wonder if you could say more about why you are so upset? I am guessing that you disapprove of all patents, or at least disapprove of defending patents. Is that right, or is there something more specific that we’ve done?

My goal in this blog was to help people see that we have a legitimate claim to infringement, under the current rules, but if you disapprove of the entire patent system, then I do understand that there’s probably nothing I can say to make you feel better about an IP lawsuit.

-- Dave Hitz

Thank you for setting Sun straight. Every large tech company should realize that the software idea patent system as currently implemented is only good for "IP" protection rackets and retaliation and is not a magic money-creation device.

Hopefully it can all be settled with some cross licensing jiggery-pokery and will not trigger the inherent MAD tendencies of the current system.

If the dust settles, it might be nice to contribute some patents to the open-source patent pool.

Apart from a gesture of good will it will save the community the work of remembering when it was "invented" last time around, which is nearly inevitable for any form of software construct these days.

I appreciate your criticism of the current patent system. Unfortunately, as you say, it is what we have, and we have to work with it for now. So you might as well use it for setting straight those who try to abuse it.

[George Mitchell]

Dave, as a long time Linux user and a fan of ZFS, I just want to thank you for being so upfront about this. As much as I value ZFS and its capabilities, I recognize that you and your company apparently have a significant investment in one of the key technologies involved. I also realize that as much I you might like to contribute those technologies to open source, you, like all the rest of us, have wages and expenses to pay, not to mention having to feed your families. I only hope that for the sake of all of the rest of us who are looking for affordable access to emerging technologies, that you can somehow reach some sort of settlement that will result in a winning situation for all of the stakeholders. I know that finding such a solution is not an easy proposition, but I wish you the best as you pursue it. At this point the ball is apparently in Sun's court. Hopefully their approach is as reasonable and understanding as is yours. In any case, no way I would compare you guys with SCO. Thanks for the integrity and sensitivity, it IS appreciated. I wish you all the best!

[Razvan Vilt]

Hi Dave,

While I am a bit amazed by the turn of things, I cannot help but wonder when will this software patent craze end? Sadly, I cannot say that I've used WAFL, but I can say that I've used ZFS, and I was impressed. I've also looked at the patent application that you filled back in '94, and while I found it impressive, I do find it to be mostly theory, and rather close to algorithmic stuff. In general, when talking about file-systems, be them WALF, NTFS, ReiserFS or EXT4, I feel that you cannot actually talk about anything but the algorithms that were designed. The algorithms make a file-system revolutionary, and not necessarily the implementation, as that is only something of quality assurance.

From my understanding, limited as it is, of the patent system and its purpose, you cannot patent an algorithm, but you can patent an implementation.

Let me make myself clear. I am a university student in IT and Electrical Engineering, and I believe that half of what our professors teach over there is available only because it cannot be patented. A quarter of the cool stuff that we were taught in university until now, and that we plan to use in our engineering work in the future, is good for patenting if you follow the _BAD_ examples that Sun, IBM, Microsoft, Oracle, NetApp provide, thus completely useless in the academic world.

Patents might protect some businesses from more or less unfair competition, but they certainly, at this rate will hinder innovation, especially at the one spot where innovation shines: in the academic level.

What would be the purpose of having a course about COW file-system design, if all the work covered is patented by one company or the other? I can't imagine anyone actually being able to use any of that information without working for the company that owns the 'Intellectual Property', or licenses it.

I certainly hope that this whole NetApp - Sun collision course can be avoided. I admire Sun for Solaris and a lot of the software surrounding-it, and I admire Sun for being the first IT MegaCorp that actually had the guts to put their money where their mouth was. IBM talks about contributions, so does HP, but most of the contributions made by companies in the wonderful FLOSS world are made by Red Hat and Sun, not by Novell, not by HP, and not by IBM.

And NO, I don't think this as something that even remotely resembles the SCO/IBM thing. You at least put everything on the web, thus we can, and will judge for ourselves.

All the best,
Razvan

P.S.: I quoted the term IP because I find-it to be one of the most hypocritical terms possible. It really is the worst choice of words possible.
P.P.S.: What I wrote above doesn't mean that IBM, HP and Novell didn't contribute, it simply states that from my POV Red Hat & Sun contributed much more, both if you count the number of lines of code and documentation, if you count the number of man-hours, and if you also take into proportion the size of the company.

[Nobody Important]

I really have no idea about the patents in question or the alleged copying, I only came because of the mention of SCO.

I'd like to point out, however, that on the issue of software patents you shouldn't expect too much sympathy. I think that a significant number of people believe that such patents should not exist at all, and I am among those. After all, all real progress is made by building upon good ideas. What would you have done if trees were patented? And I thought that copy-on-write was patented, but that may have expired by now. Not to mention that you built upon other old ideas like inodes...

Still, if you're not going to go into crazy litigation with anyone who is running ZFS (I don't run it, although I've wanted to try it), I suppose that at least is a good thing so I suppose that I shall remain grateful even for small favors.

[Charles Duffy]

As a techie with strong OSS leanings, I'm a firm believer that intellectual property law should exist with the intent and purpose of furthering the interests of the public as a whole, as suggested by the segment of the US Constitution authorizing its existence. Indeed, while our chairman of the board disagrees with me vociferously on the subject, I'm of the belief that -- were it not for defensive needs -- the startup whose product I've spent the last five years working on would be at least as well off without our patent portfolio, keeping our focus on staying ahead of our competition on the basis of our ability to improve, deliver and support our product.

That said, I can appreciate where you're at (if we were approached as you were, we'd certainly be reaching for what portfolio we have and looking at what we could leverage into a favorable cross-licensing deal -- and taking an aggressive stance if that's what we thought necessary to best represent our interests). I hope that some end can be reached which is beneficial to the public as well as yourselves. I've been keeping a watchful eye on ZFS in the hopes that it would lead to a solution to some Big Problems for companies without the money to spend 'yall's hardware. (I hope to leave this latter category in the not-too-distant future -- though when it happens, any technology purchase decisions we make /will/ be filtered through knowledge of the public actions of those companies we consider tying ourselves to through those decisions).

Frankly, if there's one end that could come out of this that would be positive for all involved, it would be a cross-licensing deal with a stipulation that Sun prominently brand and advertise ZFS as based on NetAppliance technology. You get great marketing (and gobs of goodwill with the OSS community for not killing off a well-known project), and Sun gets to keep one of the major factors differentiating OpenSolaris from Linux, even if they can't continue to take all the credit.

As IBM has found, goodwill with the OSS community does indeed translate into sales -- and getting ZFS branded as based on NetAppliance technology under terms which leave it as a viable open source project with room for adoption by others will no doubt translate into goodwill.

[Matty]

Hi Dave,

As a long time Solaris and NetApp user, I am also bothered by the fact that NetApp would choose to sue Sun over their ZFS file system. Is there a reason you and Jonathon Shwartz couldn't sit down and hash out this issue? Being in a position to recommend hardware purchases at my employer, this will definitely change my willingness to purchase anything from NetApp in the future (just as it would if Sun was suing NetApp over an open source technology). I hope you will consider the needs of your customers and the open source movement during the litigation process.

- Matty

[TimC]

Disclaimer: I make my living off of netapp staying a very heathy company...

That being said, this seems a bit off to me. First, Sun claims they were seeking cross licensing... you claim they were seeking money. This seems EXTREMELY unlikely to me. Sun isn't exactly hurting for money, nor would they have any use of *millions* of dollars from netapp. I'm skeptical at best.

Second, you decided to file this in Lufkin Texas. Best known for having absolutely 0 understanding of patents, and a haven for patent trolls...

I'm sorry but your actions do not match your words. To me it appears you fear zfs, and rather than beat them technologically, you'd prefer to do it in a court room. Protecting your technology is important, but they didn't steal your code, and quite frankly I think it's a joke to try and corner a market through the courts.

[Adam Langley]

From reading the patent [1] you appear to have "patented" functional trees (the tree structures that appear in all functional languages).

You cannot believe that this is original, so all that's left is that the application of functional trees to a filesystem. I'm strongly suspect that a motivated researcher could find prior art for this and, if not, it's certainly not "non-obvious".

So this patent would appear to be another clear abuse of the patent system.

[1] http://www.google.com/patents?id=uecCAAAAEBAJ

[Jakub Sadowski]

As one of the "technical folks" you speak of, I will first say that I'm a big fan of Sun Microsystems since they are a technology company which sets the highest standard for technology ethics. They always create a fair playing field with all the platforms and standards they develop or participate in developing. It's unfortunate and somewhat shocking to hear that they're carelessly behaving as patent leeches. After reading your clarification of this case, Dave, I can't help but agree with your course of action - patent suit or no patent suit.

[Jimmy]

Seems to me Sun was playing tackle football, and NetApp decided to pull out a live grenade. The present rules may allow it, sure. However, using existing laws to make an unjust situation even more unjust, for more people, still seems horrid.

Sun's patent abuse is despicable, NetApp's patent abuse is despicable: both are in danger of coming out of this looking like bad guys.

I don't see how a company could in good faith participate in a reform effort while pursuing one of the worst possible abuses of a software patent that justify and show the need for the very reform -- hindering the creation of new software, by attempting to say a patent gives ownership over a fairly routine algorithm for operating on a certain structure.

It's mentioned NetApp is not concerned with downloads for personal or non-commercial use, so some users of Open Source are safe.

However, this is not very comforting -- this news about NetApps' actions still has immediate and serious repurcusions for 90% of OpenSolaris users. Many, perhaps most users of Open Source software utilize the software for some commercial purpose. For instance, people run their web site on a computer system, their e-mail/database/DNS server, etc may run Open Solaris box that happens to store its system files using ZFS. Are they to be pursued next?

If future versions of the OS were to stop supporting ZFS, it is still a serious threat.

These users would have a serious problem, if all their data is stored on a ZFS storage pool, they would lose all their data and file system layout of their working system, when they later need to upgrade their kernel, for security fixes, and that is extremely harmful...

Result: Noone can really safely decide to use ZFS until this is somehow settled, it's just too risky to leave your data to chance (that a future OS upgrade may make your volume layout and saved files no-longer viable).

[joseph martins]

Dave,

I appreciate your honest and frank response, thank you.

[Robert Loomans]

Is it just me or does it seem plausible that this is one factor in Sun deciding to release ZFS as open source?

Razvan Vilt: I believe that this is the whole problem with software patents: they *are* patents on algorithms and techniques... and many feel that this differs significantly from how patent-able other areas of invention are.

[Mike]

Maybe you should open source your software stack as well, so we can see what patents you are using without paying.

Stop spreading FUD like this. Sit down with Sun and sign a patent exchange letter.

[Ken Stox]

Dave:

I must commend you on the open fashion you seem to be approaching this problem. I hope the lawyers will allow you to continue to discuss this in an open process. I don't have enough information to agree or disagree with your stance, but bringing it into the light of day speaks well for your conviction.

[Tim Vernum]

Your football analogy simply doesn't fit.

Wielding a patent arsenal is not "wearing pads and helmets". It's training your players to bust the opponents knees.
If you used your patent portfolio to force a cross-licensing agreement, then that would be "pads and helmets", but this is clearly not.

Whether you consider it to be good business, or good ethics to sue Sun is none of my concern. But this is yet another software patent dispute, and to try and frame it as something else is simply dishonest.