Comments on: The Hitz report

By: chief

chief — Fri, 22 Aug 2008 18:49:51 +0000

If ZFS is a violation of NetApp’s rights to WAFL, AMD should sue Intel for Nehalem. x86 single-die quad-core with private L2, shared L3 cache, integrated memory controller and point-to-point link? That doesn’t happen, though. And I think it highlights how silly this is.

By: Robin Harris

Robin Harris — Mon, 07 Jul 2008 01:47:01 +0000

Peter, I guess I’m missing the point or perhaps I didn’t communicate the issue I saw.

To wit: prior art doesn’t have to be “commercial” to be prior art. If Sun can point to either academic research into these techniques or, as Ryan suggested, common usage in databases, then they’ve got a case for invalidating NetApp’s patents.

I haven’t done that research, so I’m not passing judgement, despite my skepticism. What has changed with the Teleflex decision is that Sun can keep appealing the prior art issue. It used to be that once the trial court decided there was no prior art, there was no appeal. Hence the popularity of East Texas for patent cases.

Teleflex is a much tougher standard for plaintiffs, and as Ryan suggests, the techniques in question may have been widely used in related contexts. If Sun were to lose at trial, they still have a chance to persuade an appellate judge.

Thus my issue with Dave’s Bonwick quote was the qualifier “commercial.” Bonwick’s acknowledgement has no bearing on the issue of prior art.

HTH,

Robin

By: peter

peter — Sun, 06 Jul 2008 23:19:45 +0000

robin, here’s dave hitz’ complete quote of sun’s jeff bonwick (page 2 of hitz’ declaration):

“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.”

you cited only the part after the ellipses and derided hitz for trying to prove something by way of the remainder, but it’s pretty clear that hitz’ point is the part you elided.

what’s up with that?

By: Robert Milkowski

Robert Milkowski — Wed, 02 Jul 2008 23:56:43 +0000

The problem for NetApp is that si far thei WAFL was an advantage for them. Now you have ZFS with everything you got in WAFL and more and additionally you also have a support for NVRAM. Actually it is even better because you got support for NVRAM/SSD for ZIL which helps a lot for all synchronous writes but you also got L2ARC which combined with SSD gives you relatively huga cache which is much cheaper than RAM.

Then put on top of it some nice GUI, make it a black box to user, open source and…. there’s nothing to smile for NetApp.

interesting read

By: Joe Kraska

Joe Kraska — Wed, 02 Jul 2008 20:27:09 +0000

One would have to believe that if Sun were paying a reasonable license fee to NetApp the Sun Fire x4500 …
—–
Robin, this is presuming that Sun does not have an enterprise storage strategy to use ZFS for the market segment that NetApp currently services. That… presumes too much.

Mind you I agree with your conclusions regarding the patents and the like. But to say that NetApp has nothing to be afraid of is simply not true. Sun would, if they could, knock NetApp out of the storage market. And no, they’re not so stupid as to think Thumper could do that.

Joe.

By: Barry Thomas

Barry Thomas — Wed, 02 Jul 2008 17:43:07 +0000

Where can I get ZFS to try it out – ? Or get one of Sun’s “x4500″ boxes? I would love to pay less than the proprietary NAS vendors are charging – $250k for an NFS filer, what a total ripoff.

By: Shehjar Tikoo

Shehjar Tikoo — Tue, 01 Jul 2008 19:38:34 +0000

When CIO’s couldnt care less I dont think they’re taking into account the huge performance gains given to their storage infrastructure by WAFL, because by now, the low latency standards set by Netapp are generally taken for granted. But..but WAFL *is* central to high performance because WAFL is not just a filesystem but a filesystem combined with NVRAM that enables low latency request logging and delayed write gathering, two very significant factors if they’re going to be serving millions of requests per second.

From your description above, the legal issues between Sun and Netapp seem to have arisen because of the tree-of-blocks layout which has its uses in fast and low disk-space overhead snapshotting.

By: thinking sysadmin » Blog Archive » Hotlinks, 7/1/2008

thinking sysadmin » Blog Archive » Hotlinks, 7/1/2008 — Tue, 01 Jul 2008 19:08:09 +0000

[...] The Hitz report – Robin Harris at StorageMojo on the Sun-NetApp lawsuit: NetApp’s biggest misperception is that WAFL is somehow central to the success they are enjoying today. That was true about 10 years ago. Guys, your average F500 CIO today could care less about WAFL. [...]

By: Ryan Malayter

Ryan Malayter — Tue, 01 Jul 2008 13:06:54 +0000

Relational database management systems have been using the same techniques for decades to maintain transaction-consistent indexes and relational data inside databases. WAFL may indeed be the first implementation of the idea where the “tree of blocks” points to files in a general-purpose file system. But in an RDBMS, the tree of blocks (tree of index pages) typically points to arbitrary row data in the database. What’s the difference? Not much, bits are just bits. The application to a file system seems obvious to me (reasonably skilled in the art), meaning the patent should likely be challenged. Heck, not-so-innovative Microsoft was kicking the same ideas around back in the early 1990s as part of the WinFS file system for the “Chicago” project.