NetApp – not Oracle – announced that
. . . both parties have agreed to dismiss their pending patent litigation, which began in 2007 between Sun Microsystems and NetApp. Oracle and NetApp seek to have the lawsuits dismissed without prejudice. The terms of the agreement are confidential.
Inquiring minds might wonder what “without prejudice” means. The helpful Business Dictionary says
When used in a document or letter, these words mean that what follows cannot (a) be used as an evidence in a court case, (b) be taken as the signatory’s last word on the subject matter, and (c) be used as a precedence [sic].
Technically NetApp is reserving the right to refile the suits should, say, Oracle makes them angry. But the probable meaning is that NetApp simply didn’t want to admit legally that their patents didn’t hold water.
The StorageMojo take
While I’m pleased by this – props to CEO Georgens, who supported this decision – I’m not surprised. As I noted a year ago:
. . . Oracle is in a stronger position to negotiate a settlement with NetApp over the ZFS/WAFL patent suits. After all, why would a storage company want the world’s largest database company as an enemy?
Answer: no reason at all. But why did it take this long for common sense to prevail? Somebody powerful inside NetApp needed to be clubbed into submission first.
Sure, Exadata makes Oracle run fast. But Oracle also sells business apps and middleware that don’t need screaming performance – and Sun isn’t strong in NFS and CIFS storage.
NetApp is one of Oracle’s dozen Global Partners. Much better than being one of Oracle’s Global Enemies.
Courteous comments welcome, of course. BTW, Oracle webmaster, it’s safe to take Sun off the list of Global Partners.
Robin,
This is all well and good for Oracle but NetApp hasn’t rescinded the threat to Coraid and therefore the rest of the ZFS ecosystem.
Someone more devious than I could say NetApp wants to win a suit or settlement against a little guy to use as precedent.
NetApp: Come right out and say ZFS is OK.
This may make business sense, but it would have been nice to settle the question of the validity of NetApp’s patent that they claimed ZFS violated.
I believe some have been invalidated, but given that ZFS is open source and used elsewhere (e.g., FreeBSD, Nexenta), a conclusive answer to the all of them is desirable for some of us.
The Palindromic Times headline reads:
‘s Poop patent! NetApp Oops
In FreeBSD Project the ZFS is alive 😀
http://www.freebsdnews.net/2010/09/03/freebsd-continue-supporting-zfs/
Please rewrite this article and add this link!
Thanks
Carlos Jardim (Brazil)
My 2 cents is that Larry Ellison was tired of NetApp getting a lot of free publicity at Oracle’s expense and sent someone over who asked bluntly what NetApp really wanted. NetApp replied, “We want to be ZFS’s Daddy.”. Oracle said, “Fine. Drop the lawsuit and we will work out the details with you”. This does not bode well for all the ZFS lovers out there. NetApp desperately needs a WAFL replacement or infusion. ZFS would be ideal. Oracle has no plans for it except to allow it to wither away in “benign neglect”. Oracle could have M&A plans for NetApp as soon as it finishes digesting Sun. Oracle did not buy Sun for its Storage capability.
Too little, too late. How many people honestly care about ZFS anymore? Don’t get me wrong I love the idea of ZFS, and I wish it had been licensed so that it could be in Linux.
But as it is Solaris is going to be integrated with Oracle products, OpenSolaris is dead, and while FreeBSD has ZFS, the general consensus is that it does not have the same performance and stability as ZFS on Solaris.
So unless you plan on buying a “total solution” from Oracle I doubt you will see ZFS in a production environment near you. And the geek in me cries a little at that thought…
I disagree with Robert comments. Far fetch in my book. I do not think NetApp need ZFS neither Oracle will drop it. I’m sure ZFS is a key element for Oracle today. ZFS start to mature (less bugs and stable features/functions). Like any file systems they mature after 10 years or so. The interest on ZFS remain high. More than other older file systems like Veritas, UFS, NTFS for example.
ZFS need to scale across nodes like NetApp ONtap 8.0 does. It is not there yet and does not contains code to do it. I think Sun bought Lustre in part to acquire the scalability and code needed to do it. I can see a merge between ZFS and Lustre as well as SAMFS within 5 years. Oracle own several file systems (ZFS, SAMFS, Lustre, Celeste, Honeycomb, NFS few others) and need to reduce them to cut cost.
Oracle bought Sun in part for its OS and ZFS is part of it by default today…not UFS. They bought them also to have the entire stack and control all elements that can stream their databases.
The settlement might be a step to announce new products that leverage ZFS…or NetApp…or both in different products at Oracle World next week. Who knows…
RE: “I can see a merge between ZFS and Lustre as well as SAMFS within 5 years.”
Sun bought LSC Inc. in 2001 which gave them SamFS-QFS. If you add your 5 year time to that 10 years that is 15 years to integrate and develop a mature product. Too long. Even taking into account ZFS and Lustre. The economic downturn is the only thing that saved them.
The rumor at the time was that Sun did not want SamFS, only QFS. At that time you could deliver a working “Point Product” solution using Veritas VxVM, HighGround SRM and SamFS, which included QFS but that presented more problems than help. People were reluctant to change UFS for QFS.
Last thought, Larry Ellison is a business man, not a “techie”. So are the guys at NetApp. Both are excellent at marketing. Both have gotten excellent mileage out of legacy technology with excellent “marketecture”. Both are now behind the curve. Who is Oracle’s number one competitor? IBM. How does IBM do the database technology? Does it give them the competitive edge?
The name of the game is replication. Instantaneous replication without data corruption. Database corruption can be like a wildfire. You need to eliminate all the corruption sources you can. There is no time to come back from disk and certainly not tape. Web front ends are the same. You need a Flash Matmos layer…
What does this mean (if anything) for ZFS and Apple?
Cheers
Scott,
Well, I *hope* it means that Apple gets back to replacing the decrepit HFS+ with ZFS as they’d planned. But I suspect that the engineering team has a a plan they’re loathe to give up. A phone call from Steve to NetApp CEO Tom Georgens is probably all it would take to get NetApp on board, and since Larry and Steve are buddies, it’s a done deal.
But Steve isn’t a plumbing guy, so that is a stretch. On the other hand he wasn’t an operations guy at one time either, and he figured that out when he needed to. Maybe he’ll surprise us all again.
Robin
I wonder how ZFS fits with these plans?
“Is EMC the Next Step in Oracle’s Journey to $100B?”