Slashdot Mirror

On the "Metro" thing the first thing that came to mind right now was the NetApp MetroCluster we just installed at work.
Now I find there is an IBM storage technology called MetroCluster and a NetApp storage technology called MetroCluster

I wonder how that will pan out.

Virtualization will not isolate them against each other. For example, it is quite easy to saturate I/O from the playground.

That is an architecture issue. Implement Vsphere Enterprise+ with Network I/O control, Storage I/O control.

Put the playground on different SAN LUNs from the production LUNs.

Place the playground LUNs backed by different physical disks on separate vFilers, and/or use FlexShare to prioritize production workloads.

Leverage vShield App / vCloud networking and security, to ensure IT playgrounds don't have internet access, or the ability to export data or be used as a covert channel to escape the network; that they run on a separate vSwitch.

If you're really paranoid, use direct-mapped crypto hardware or CPU affinities to defend against academic timing attacks (at the cost of scalability and system performance)

I noticed the summary conveniently forgets to mention that there are also several hundred boxes of paper evidence.

When you already own entire warehouses to hold evidence, storing a few hundred boxes of paper is not expensive

And of course it isn't like these are 2 terabytes of Blu-ray movies: it's probably mostly text and image files, and that is a lot of text documents to keep track of and make sure are backed up on a regular basis, with a full chain of custody to ensure they aren't being tampered with and whatnot. Sure, 40 TB sounds like a small amount of data, but then again if you introduce 4 or 5 backups with tampering resistance... it suddenly starts looking like quite a bit.

Any of the major storage vendors will be happy to sell you a WORM storage array that prevents tampering and has remote replication.

http://www.emc-centera.com/more-about-emc-centera.htm
https://communities.netapp.com/community/products_and_solutions/netapp_integrated_data_protection/blog/2011/12/19/netapp-snaplock-where-compliance-and-efficiency-meet

FYI: Net Applications =/= NetApp (formerly known as Network Appliance)

NetAPP's product is not RAID6; it's RAID-DP. What they are implementing is definitely not RAID6,
specifically they have implemented a proprietary scheme, and it is only marginally less performant than RAID10.

I stand behind my original statement that you should not discount RAID5, RAID6, SATA etc - it is the end metrics that matter.

I don't want to bore you by arguing the same points... but you did specifically draw attention to RAID-DP with a bold not plus an incorrect salesdroid assumption. So I'll serve you up a hot slice of humble pie; directly from NetApp's website:

"RAID-DP is a Double Parity RAID 6 implementation that prevents data loss when two drives fail."

http://www.netapp.com/us/products/platform-os/raid-dp.html

I know how RAID-DP differs; but fundamentally it is RAID-6. I am familiar with SAN and NAS on a technical level.

Color me surprised if Netapps actually use hardware raid, since everything else is handled by the software, according to http://en.wikipedia.org/wiki/NetApp_filer and articles like http://media.netapp.com/documents/tr-3298.pdf

Obviously they have non-commodity controllers, but are these "hardware raid"? So far, nothing indicates that they are from cursory reading. If you have links that say that they are, do share.

But you sound like an old storage guy, so it's obvious why you like hardware raid. But it's a brave new world out there my friend.

Agreed. I think the shift has occurred because of increasing corporate interest in open source. BSD is seen as more corporate-friendly than GPL, when in fact it should be the other way around--BSD allows your competitors to reap the fruit of your labor without giving you anything in return. Start-ups, however, are lured by the idea of being able to close-source everything once their product becomes a smash hit,

There may well be start-ups who are. There are other start-ups who incorporate BSD-licensed code in their software and never open-source their software in the first place. One such startup succeeded rather well.

while established companies face genuine legal issues preventing them from linking GPL'ed code with closed-source code from vendors.

Unless it's an established company that doesn't link GPL'ed code with closed-source code.

Speaking of EtherDrives, it appears from Wikipedia that they don't connect using SMB, or anything that is TCP/IP based. They use another protocol called ATA over Ethernet.

Not sure about Coraid's or NetApp's line of products, though.

http://blogs.netapp.com/dave/2007/09/netapp-sues-sun.html The victim has become the offender.

ZFS is not encumbered by asshattery any more than Linux is.

It's encumbered by a patent dispute

While Linux is encumbered by the SCO mess, which is basically the same thing, except related to copyrights and ownership of codes, rather than ownership of the whole concept of copy on write.

This kind of thing is fairly big in some circles.

It's interesting what Schwartz has to say about how things work "on the inside". Companies bluffing and calling each other's bluff. Showing up and going "I'm watching you". His description makes it sound a bit like Jobs & Gates hadn't really thought their cunning plan all the way through, which I would think is unlikely. I'd have guessed they were just testing Sun's resolve, finding out how Sun evaluated their own patent portfolio, investigating whether these projects (Looking Glass and OpenOffice) were just a tech demo or were something that Sun wanted to stand by and protect. What his blog post didn't mention was on how many occasions Sun did the same thing to another company, big or small. It would be laudable if they refused to do that but it would also mean they were deliberately pulling their punches, so it would be a bit surprising from a large corporation.

NetApp sued sun over patents ZFS arguably violated: http://www.sun.com/lawsuit/zfs/. But NetApp alleged that Sun had first demanded patent royalties from NetApp and that they were acting in response to that: http://blogs.netapp.com/dave/2007/09/sun-patent-team.html

Who knows where the truth lies over the ZFS case but it does open the prospect that Sun wasn't sitting passively by and getting threatened by other companies. On the other hand, there could be more to this story than meets the eye (e.g. the kind of high level meetings Schwartz refers to, preceeding the legal letters) in which case it might not be anything like so simple. We've not generally seen Sun visibly holding back (or trying to) the marketplace using patents as much as, say, MS or Apple might have done. But it doesn't mean that given their investment in patents they didn't try to use them.

I would suggest something like this:
http://www.netapp.com/us/products/storage-systems/fas2000/fas2000-tech-specs.html

that is, if you have a datacenter and ability to get 20A (NEMA L6-20) or 30A (NEMA L6-30) power.

Are the datasets generated using similar chunks of data? If so, deduplication could be very very helpful (in addition to compression)

TCP overhead at 1GbE for a modern processor is negligible - you're only talking about processing 120MB/sec or so.

Here is a document including a pretty graph: http://media.netapp.com/documents/tr-3628.pdf

"...enabling the TCP Offload Engine (TOE) on the Linux hosts did not noticeably affect performance on the IBM blade side."

Considering what's going on between NetApp and Sun currently, I wonder what they'll think of this?

-yb

Considering what's going on between NetApp and Sun currently, I wonder what they'll think of this?

-yb

Re: That's true for any licensee - in fact, Net App could adopt ZFS today and receive the same protection. The port is done to FreeBSD, the OS on which Net App's filers are built.

Last I checked NetApp was using some NetBSD-derived components in their Filer OS.

Check more recently, please.

OK, to be fair, ONTAP Classic, while it has a fair bit of BSD code in it (some commands, the networking stack, etc.) isn't built on anything that matches "*BSD", so the person to whom you're replying is overstating the case. ONTAP GX, however, is FreeBSD-based.

Perhaps you're confusing NetBSD with FreeBSD ?

To quote, "Interestingly, our advanced ONTAP GX architecture is built on top of a full UNIX release. We took Data ONTAP, including WAFL and RAID, combined it with the new code from our Spinnaker acquisition, and hosted the combined result on FreeBSD in a combination of user processes and kernel modules. For security and simplicity we have disabled and hidden many parts of FreeBSD." : http://blogs.netapp.com/dave/2007/04/is_data_ontap_b.html

So Sun gets to decide what SSD means?

In the context of an example using Sun and you saying it's incorrect because of semantic irrelevancies, yes.

Thumb drives are about $2/gig on the low end, while SSD's are about $3/gig on the high end.

I think you mean "low end". An 80GB Intel X25-M ("low end") is around $280 (so >50% more than a good USB thumbdrive). A 32GB Intel X25-E ("mid range") is about $380 (~$11/GB). A "high end" SSD (as you might find in an EMC SAN) probably costs 2-3x that.

Now you are stearing away from the posters point completely. His point is that SSD's are replacing rotational media, and they are.

Not when storage volume is a factor, they're not. If you want a dozen terabytes of disk space, SSDs are almost certainly not a realistic option.

Hard Drives didnt supercede them, so I guess your point is that hard drives are a cache for 2" tape reels, that SSD's are a cache for hard drvies, and RAM is a cache for SSD's?

These are typically referred to as HSM systems. Although so-called "MAID" is beginning to displace tape in many installations as the penultimate tier.

However, my actual point is that multilevel caching exists everywhere in computing, and has done pretty much forever, so attacking it as a principle by insisting a faster storage tier is going to just flat-out replace another is a bit silly. SSDs (or "flash", if you prefer) are going to find just as much - if not more - usage as an extra caching layer, as they will for dedicated storage. NetApp will almost certainly be rolling out an SSD-based complement to their existing DRAM-based Performance Acceleration Module in the near future. I expect an SSD-based Accelerator Appliance will be coming as well.

Heck, with ZFS you can even DIY an SSD-accelerated storage system. and get nearly all the performance benefits for relatively little cost increase.

The fact is that for all intents and purposes, SSD's will be replacing hard drives.

It will be quite a while before SSDs replace multi-terabyte disk arrays wholesale, especially when using them as a caching layer can deliver 90% of the performance benefits at a mere fraction of the cost (this is the principle Sun's 7000 series are built around - cheap and big SATA disks on the back end, with a much smaller SSD cache sitting in front of it). The same applies to consumer usage - a 1TB drive with a well managed 10-20GB SSD cache in front of it will deliver nearly all the benefits of an SSD, without massively increasing price or sacrificing capacity and usability.

If all your data will fit into the puny sizes of current SSDs, great. I suspect most people's will not, especially in these days of HD handycams and movie downloads.

So Sun gets to decide what SSD means?

In the context of an example using Sun and you saying it's incorrect because of semantic irrelevancies, yes.

Thumb drives are about $2/gig on the low end, while SSD's are about $3/gig on the high end.

I think you mean "low end". An 80GB Intel X25-M ("low end") is around $280 (so >50% more than a good USB thumbdrive). A 32GB Intel X25-E ("mid range") is about $380 (~$11/GB). A "high end" SSD (as you might find in an EMC SAN) probably costs 2-3x that.

Now you are stearing away from the posters point completely. His point is that SSD's are replacing rotational media, and they are.

Not when storage volume is a factor, they're not. If you want a dozen terabytes of disk space, SSDs are almost certainly not a realistic option.

Hard Drives didnt supercede them, so I guess your point is that hard drives are a cache for 2" tape reels, that SSD's are a cache for hard drvies, and RAM is a cache for SSD's?

These are typically referred to as HSM systems. Although so-called "MAID" is beginning to displace tape in many installations as the penultimate tier.

However, my actual point is that multilevel caching exists everywhere in computing, and has done pretty much forever, so attacking it as a principle by insisting a faster storage tier is going to just flat-out replace another is a bit silly. SSDs (or "flash", if you prefer) are going to find just as much - if not more - usage as an extra caching layer, as they will for dedicated storage. NetApp will almost certainly be rolling out an SSD-based complement to their existing DRAM-based Performance Acceleration Module in the near future. I expect an SSD-based Accelerator Appliance will be coming as well.

Heck, with ZFS you can even DIY an SSD-accelerated storage system. and get nearly all the performance benefits for relatively little cost increase.

The fact is that for all intents and purposes, SSD's will be replacing hard drives.

It will be quite a while before SSDs replace multi-terabyte disk arrays wholesale, especially when using them as a caching layer can deliver 90% of the performance benefits at a mere fraction of the cost (this is the principle Sun's 7000 series are built around - cheap and big SATA disks on the back end, with a much smaller SSD cache sitting in front of it). The same applies to consumer usage - a 1TB drive with a well managed 10-20GB SSD cache in front of it will deliver nearly all the benefits of an SSD, without massively increasing price or sacrificing capacity and usability.

If all your data will fit into the puny sizes of current SSDs, great. I suspect most people's will not, especially in these days of HD handycams and movie downloads.

The NetApp vs Sun lawsuit over ZFS isn't going the way NetApp would like it to ...

http://www.sun.com/lawsuit/zfs/index.jsp

To the contrary, NetApp may end up like SCO vs Novell, where the initial complainant ends up owing the respondent. Sun could very well end up both pwning AND owning NetApp.

As for the antivirus companies - I wish, but there will always be *some* "useful fools" around, and people whose financial self-interest aligns with enabling them to stay dumb and foolish.

According to a post on a Sun website

And, guess what, if you look on NetApp's website they disagree!

http://blogs.netapp.com/dave/2008/11/lawsuits-and-fo.html

Neither of these are likely to be good sources of information. It's quite possible that prior art will deal with all the patents in play by both companies.

Pete

If performance is your goal, don't look at the "buy-it-at-Frys" level of NAS, or at "roll-your-own."

If you build your own, you'll end up bottle-necked by the performance of the particular OS you use, plus SAMBA or NFS (depending on your needs.) Plus, there's the time factor in putting it together, tuning it, and maintaining it. Granted, this isn't a lot if you're already a tech-head, but your time isn't free.

If you buy one of the consumer-level NAS boxes, what you're getting is the equivalent of building your own, without the ability to tune it, since most are based on the same open-source software you would use yourself. Pretty much every NAS device I've ever seen has the same cyclical bursty transfer profile as a build-it-yourself.

If you want better performance and buzzwords, every major PC vendor now has a SAN solution. You get the benefit of a team of people whose job it is to maximize disk performance, and a nice management system. However, your system head still has the same problems as before - using a general purpose OS and/or open-source software.

If you want pure performance, look at Network Appliance. They've been in the game for a long time, and their hardware/software combination allows them to control/tune the whole environment. To a first approximation, all the cool things in ZFS were done ten years ago by NetApp. You get the benefit of a whole company whose job it is to maximize disk and network performance. You can look at a performance review from earlier this year showing about 30k SPC-1 IOPS.

Personal anecdotes:

1. Several years ago, we did a benchmark for ClearCase between a Sun hardware head with a (a) directly connected, fibrechannel SCSI RAID array, and (b) a 100G ethernet connection to a NetApp. The performance of the NetApp was about 20% higher than directly connected disk.
2. NetApp service is incredible. We came in one morning, and there was an email from Network Appliance that basically said "Hello; your NetApp notified us that one of its disks has failed. We have shipped a replacement. Here's your UPS tracking number."
3. The above also holds for software. There's nothing like "Hello; your NetApp had a software failure. From analyzing the crash dump it sent us, we recommend you install patch xxyyzzz."

Note: My only relation with NetApp is being a very satisfied customer.

If performance is your goal, don't look at the "buy-it-at-Frys" level of NAS, or at "roll-your-own."

If you build your own, you'll end up bottle-necked by the performance of the particular OS you use, plus SAMBA or NFS (depending on your needs.) Plus, there's the time factor in putting it together, tuning it, and maintaining it. Granted, this isn't a lot if you're already a tech-head, but your time isn't free.

If you buy one of the consumer-level NAS boxes, what you're getting is the equivalent of building your own, without the ability to tune it, since most are based on the same open-source software you would use yourself. Pretty much every NAS device I've ever seen has the same cyclical bursty transfer profile as a build-it-yourself.

If you want better performance and buzzwords, every major PC vendor now has a SAN solution. You get the benefit of a team of people whose job it is to maximize disk performance, and a nice management system. However, your system head still has the same problems as before - using a general purpose OS and/or open-source software.

If you want pure performance, look at Network Appliance. They've been in the game for a long time, and their hardware/software combination allows them to control/tune the whole environment. To a first approximation, all the cool things in ZFS were done ten years ago by NetApp. You get the benefit of a whole company whose job it is to maximize disk and network performance. You can look at a performance review from earlier this year showing about 30k SPC-1 IOPS.

Personal anecdotes:

1. Several years ago, we did a benchmark for ClearCase between a Sun hardware head with a (a) directly connected, fibrechannel SCSI RAID array, and (b) a 100G ethernet connection to a NetApp. The performance of the NetApp was about 20% higher than directly connected disk.
2. NetApp service is incredible. We came in one morning, and there was an email from Network Appliance that basically said "Hello; your NetApp notified us that one of its disks has failed. We have shipped a replacement. Here's your UPS tracking number."
3. The above also holds for software. There's nothing like "Hello; your NetApp had a software failure. From analyzing the crash dump it sent us, we recommend you install patch xxyyzzz."

Note: My only relation with NetApp is being a very satisfied customer.

Interestingly, our advanced ONTAP GX architecture is built on top of a full UNIX release. We took Data ONTAP, including WAFL and RAID, combined it with the new code from our Spinnaker acquisition, and hosted the combined result on FreeBSD in a combination of user processes and kernel modules. For security and simplicity we have disabled and hidden many parts of FreeBSD.

The older codebase was originally derived from the original BSD Net/2 release:

The first version of Data ONTAP borrowed lots of code from Berkeley Net/2 (one of the earliest open-source releases of UNIX), including the TCP/IP stack, system boot code, and device drivers. Since then, we've borrowed liberally from other open-source UNIX releases. We wrote the command line interface from scratch, but we designed it to look like UNIX, since our first market was UNIX system administrators. Clearly, ONTAP is related to UNIX.

The full post is worth reading - he talks about the relationship between ONTAP and UNIX in some detail.

Dave, the CEO of NetApp had a very interesting blog posting about patents last summer. Especially interesting since NetApp and Sun are now in a Patent lawsuit battle royale.

This is why SAN manufacturers have come up with "thin provisioning". NetApp is quite good it, read more here.

Use NetApp.

It's hands down the best storage platform for virtualization technologies. It slays any storage vendor I've seen when it comes to backing up and restoring virtualized environments, not to mention thin provisioning, deduplication, and instant cloning of datastores without taking up any space. Imagine being able to backup an entire VMware farm in seconds, and streaming to tape (if you must) during the day. Any other array seems to be dumb storage that doesn't do squat to help you with your backups.

Check here: http://www.netapp.com/go/techontap/matl/VMworld07.html

They also have this cool video showing them provision 100 VMs (ready to be powered on) in 13 minutes, while only taking up 7GB. Jaw dropping to say the least: http://www.netapp.com/go/techontap/matl/downloads/VDI/Flash/VDI.html

Use NetApp.

It's hands down the best storage platform for virtualization technologies. It slays any storage vendor I've seen when it comes to backing up and restoring virtualized environments, not to mention thin provisioning, deduplication, and instant cloning of datastores without taking up any space. Imagine being able to backup an entire VMware farm in seconds, and streaming to tape (if you must) during the day. Any other array seems to be dumb storage that doesn't do squat to help you with your backups.

Check here: http://www.netapp.com/go/techontap/matl/VMworld07.html

They also have this cool video showing them provision 100 VMs (ready to be powered on) in 13 minutes, while only taking up 7GB. Jaw dropping to say the least: http://www.netapp.com/go/techontap/matl/downloads/VDI/Flash/VDI.html

According to this netapp paper even NFS over 10GbE is lower latency than 4Gb FC. I imagine if the processing overhead isn't too high or offload cards become available then this would be significantly faster than 4Gb FC. As far as bandwidth 10>4 even with the overhead of ethernet framing, especially if you can stand the latency of packing two or more FC frames into an ethernet jumbo frames.

The port is done to FreeBSD, the OS on which Net App's filers are built.

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."
http://blogs.netapp.com/dave/2007/09/netapp-sues-sun.html

Here's NetApps CEO's blog post about this.

Here is an interesting link...

http://www.netapp.com/go/Sun%20Lawyer%20Email.pdf

The new data ONTAP GX is based on FreeBSD.
http://blogs.netapp.com/dave/2007/04/index.html

Have a look at http://blogs.netapp.com/dave/2007/09/litigoperatio n-.html before jumping into any conclusions.

Hitz is a smart and technical person. He is not a "suit", he's very honest and says things as they are.

Look at the specifics of the ZFS paper, and look at the specifics of the
WAFL patents.

Hope you win NetApp.

From: http://blogs.netapp.com/dave/

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.

Apparently, it's more than that. Someone upthread pointed out this blog post by one of the NetApp staff. ZFS doesn't just implement the same features as NetApp's WAFL file system, the way it implements them is the same too, and Sun staff have said that they were more than a little inspired by it.

NetApp has been around for a very loooooooong time. Almost as long as Sun.

If "almost as long as" is defined as "about 2/3s as long as". Sun: 25 years. NetApp: 15 years.

Someone has the patent on the concept of managing 'something' with 'something'? In this case files with a filesystem?

No. Go read NetApp's complaint, which enumerates the patents Sun claims NetApp is infringing and NetApp claims Sun is infringing, then look up the patents at the US Patent and Trademark Office patent search-by-number page.

And here is NetApp's boss blog

Dave's one of the founders, but he's not the CEO or president, he's an executive vice president, as per the NetApp executive biographies page.

Its interesting to read one of his other blog postings that he points out in this blog posting about how he thinks the patent system is broken, and its main goal is to force everyone to cross license patents...

But I'm torn.. I've been looking at NAS,SAN boxes, mainly the StoreVault S500, or the Higher End Netapp 270, or a lower end Sun StorageTech 52xx for my work.. I hate patents, love ZFS, but not sure which one to order now! Guess I'll have to give Equallogic another call..

Yes, cross-licensing fees: your money plus access to your patent portfolio.

http://blogs.netapp.com/dave/2007/06/how_the_paten t_.html

BigCo wants your money, but also access to your patents, so the cross-licensing negotiation begins. You try to identify patents that cover their products, and they try to identify patents that cover yours. In the end, you balance how much of your revenue their patents cover, and how much of their revenue your patents cover. The little guy usually loses. Ironically, patents on your super-special technology often don't cover any of BigCo's products, so "miscellaneous" patents may be more important than the ones you thought were so valuable. BigCo has so many patents that it's hard to examine them all, so they will do a statistical analysis showing how many patents they have that might apply to you. Why not save the lawyers fees and just assume that some percentage does apply? (NetApp was in a cross-licensing negotiation with a large semiconductor company, and since we don't design or manufacture chips, we got them to remove those patents from their analysis.)

Again, from NetApp's boss blog: http://blogs.netapp.com/dave/2007/06/how_the_paten t_.html (I really like this guy, having read 2 articles):

Don't patents ever protect your good ideas?! In theory, they should, but in practice it doesn't usually work that way. Suppose your small company wants to protect its ideas against a big company. Filing suit will accelerate the cross-licensing negotiation, and you'll probably end up paying. Better to let sleeping dogs lie. Patent battles between small companies work poorly because they are so expensive and take so long. Better to fight it out in the marketplace. Big companies suing small companies have a harder time than you'd imagine, because the courts recognize that an incorrect decision, especially against a startup, can cause irreversible damage. Courts are reluctant to impose injunctions, even if the patents really do apply. In the case of two big companies, both almost always violate each other's patents, so they end up cross licensing. (I'm not saying that patents never help to protect good ideas. If someone steals your patented idea, it's perfectly reasonable to go after them. I'm just saying that it seldom works out as well as you might hope.)

I know that some people are so frustrated with the patent system that they want nothing to do with it. The problem is, there's no good way to opt-out.

Here is the original complaint (PDF): http://www.netapp.com/go/ipsuit/spider-complaint.p df .
And here is NetApp's boss blog: http://blogs.netapp.com/dave/ (quoted below):

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.

Here is the original complaint (PDF): http://www.netapp.com/go/ipsuit/spider-complaint.p df .
And here is NetApp's boss blog: http://blogs.netapp.com/dave/ (quoted below):

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.

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?

1. Tape drives - tape capacities are universally quoted in base-10.
2. Hard disk sectors
   1. While the payload is often a power of two, the actual sector on the disk consists of many more bits - error correction, addressing, etc, that do not sum up to a power of two.
   2. Most 'enterprise grade' raid systems (HP, EMC, Netapp, etc) use drives with sector payload sizes of 520 bytes, not 512.
3. Total capacities of 'enterprise grade" raid systems.
4. Flash drives of all forms - contrary to your claim, a 2GB compact flash drive is typically 2*10^9 bytes.

if all blocks become allocated, it becomes impossible to unlink(2) a file.

The price of copy-on-write. Other systems with a copy-on-write file system can exhibit this behavior.

With drive sizes of today, and failure rates... For my money its only RAID 6.

Pros:
You can used mismatched drive sizes as long as the largest drive is your parity(s).
You can add drives up to your raid group size.
Using RAID 6 you have diagonal/horizontal parity and can recover from a double disk failure.

Cons:
Without a fancy controller speed can be an issue.

Here is some more info...

http://www.tomshardware.com/2006/01/02/safer_6_for _raid_controllers/
http://www.netapp.com/ftp/netapp-raid-dp.pdf

(I just like the chart in the netapp info.)

A double disk failure may be very unlikely, but a disk failure combined with a read-error during rebuild isn't...

If the read-error is recoverable, then there is no problem. And if it is not, then you contradict yourself, for such an unrecoverable read-error would be a disk-failure, which you agreed is very unlikely to overlap the replacement process for an earlier failed disk.

Your link (and I'm rather hesitant to trust a storage vendor's advice to buy more storage, BTW) is not explicit — could it imply an undetected read error? These could happen any time (RAID or not, array-rebuild or not), and there ought to be means of detecting them...