[9fans] OT: ZFS

[9fans] OT: ZFS

[Sam]

http://www.sun.com/2004-0914/feature/?biga=15

Parts of this article are downright hilarious:

Logically, the next question is if ZFS' 128 bits is enough. According to 
Bonwick, it has to be. "Populating 128-bit file systems would exceed the 
quantum limits of earth-based storage. You couldn't fill a 128-bit storage 
pool without boiling the oceans."

... who wouldn't want to boil the oceans?!  I mean, c'mon -

Sam

Re: [9fans] OT: ZFS

[boyd, rounin]

> ... who wouldn't want to boil the oceans?!  I mean, c'mon -

who needs to?  we've already got global warming, just let it ramp up a bit ;)

Re: [9fans] OT: ZFS

[Dan Cross]

Sam <sah@softcardsystems.com> writes:
> ... who wouldn't want to boil the oceans?!  I mean, c'mon -

Hmm, there are a lot of lobster's in the ocean.  I'll bring the
clarified butter.

	- Dan C.

Re: [9fans] OT: ZFS

[Tim Newsham]

> http://www.sun.com/2004-0914/feature/?biga=15

    ZFS is supported on both SPARC and x86 platforms. More important, ZFS
    is endian-neutral. You can easily move disks from a SPARC server to an
    x86 server. Neither architecture pays a byte-swapping tax due to Sun's
    patent-pending "adaptive endian-ness" technology, which is unique to
    ZFS.

Are they patenting receiver-makes-right?

> Sam

Tim N.

Re: [9fans] OT: ZFS

[geoff]

adaptive endian-ness is pretty easy; I nearly put it into
Ken's file server a few weeks ago and am now motivated to do it
to deny Sun the ability to claim that only ZFS does it.
I don't know how Sun does it, but you can do it pretty simply:
at a known offset in the first superblock of a file system,
mkfs or equivalent writes a long or vlong containing a pattern
like 0x01020304 (if the superblock already contains a suitable
magic number, that can be used instead).  When mounting the file
system, if that long looks like 0x04030201 instead, you push
the existing "x" byte-swapping pseudo file-system that knows the format
of all meta-data and byte-swaps blocks as they are read and written.

I believe that "x" was added to let x86 file servers serve
from disks written by MIPS file servers.

2^128 bytes really is vast.  Even 2^64 bytes is an awful lot
of data.  I'm sure that a few people can find a use for such
large files and a small fraction of those people probably even have
legitimate uses for them.  I think that most people will find
64-bit file servers to be adequate for quite a while (though
obviously Microsoft will continue to push hard here).  Unless
you're making clever use of sparse files, you'll have to wait
until you get afford 8 exabytes (it's actually 63-bit file
sizes if off_t or equivalent is signed) of storage (that's
9,223,372,036,854,775,808 - 8 binary exabytes - of disk, not
8 video tapes) before you can make use of file sizes larger
than 63 bits.

Re: [9fans] OT: ZFS

[Lucio De Re]

> I believe that "x" was added to let x86 file servers serve
> from disks written by MIPS file servers.

Yep, NetBSD has had the necessary feature for a while already, I'm not
sure if it's auto-detecting, but I think it is.

++L

Re: [9fans] OT: ZFS

[rog]

> adaptive endian-ness is pretty easy; I nearly put it into
> Ken's file server a few weeks ago
[...]
> When mounting the file
> system, if that long looks like 0x04030201 instead, you push
> the existing "x" byte-swapping pseudo file-system that knows the format
> of all meta-data and byte-swaps blocks as they are read and written.

i'm not sure that this is the kind of thing that sun is doing.  in
that press release they insist that "neither architecture pays a
byte-swapping tax" (which is what you're paying by using an extra
level of indirection).

ahh, i suppose they probably mean "neither architecture pays a
byte-swapping tax *on a filesystem created by that architecture*", in
which case you're doing something equivalent.

could they really patent such a trivial thing?

Re: [9fans] OT: ZFS

[Fco. J. Ballesteros]

> that press release they insist that "neither architecture pays a
> byte-swapping tax" (which is what you're paying by using an extra
> level of indirection).

Someone has to swap them, if they don't use the same order.

> ahh, i suppose they probably mean "neither architecture pays a
> byte-swapping tax *on a filesystem created by that architecture*", in
> which case you're doing something equivalent.

I think it has to be the same thing, or a similar trick.

> could they really patent such a trivial thing?

Well, I think there are patents on even more simple things.

Re: [9fans] OT: ZFS

[Nigel Roles]

>>ahh, i suppose they probably mean "neither architecture pays a
>>byte-swapping tax *on a filesystem created by that architecture*", in
>>which case you're doing something equivalent.
>>    
>>
We think that they mean "both architecture are taxed the same". With all 
that
space in their 128 bit filesystem, they can store all the offsets in ASCII.

Re: [9fans] OT: ZFS

[Tim Newsham]

> Someone has to swap them, if they don't use the same order.

Not really..  Could be ascii.  xml even.  ;-)

Tim N.

Re: [9fans] OT: ZFS

[Ronald G. Minnich]

On Fri, 17 Sep 2004 rog@vitanuova.com wrote:

> could they really patent such a trivial thing

Hey, it's the US. The old saying used to be that a clever prosecutor could 
get a grand jury to indict a ham sandwich. Now you can patent it.

ron

Re: [9fans] OT: ZFS

[Bruce Ellis]

i knew the 128 bit registers and instructions in the ps2
would come in handy for something absurd.  actually ZFS
doesn't go close to journalfs where all quantities are
arbitrary length.  i don't intend to add mpseek() in
case i need it, nor mpstat() etc.

brucee

> 2^128 bytes really is vast.  Even 2^64 bytes is an awful lot
> of data.

Re: [9fans] OT: ZFS

[YAMANASHI Takeshi]

On Thu Sep 16 04:05:45 JST 2004, Dan Cross wrote:
> Hmm, there are a lot of lobster's in the ocean.  I'll bring the
> clarified butter.

Hey, wait!!  I want to eat fish raw!!
-- 

Re: [9fans] OT: ZFS

[boyd, rounin]

> Hey, wait!!  I want to eat fish raw!!

lobster sashimi is quite special.

Re: [9fans] OT: ZFS

[Dan Cross]

YAMANASHI Takeshi <uncover@beat.cc.titech.ac.jp> writes:
> On Thu Sep 16 04:05:45 JST 2004, Dan Cross wrote:
> > Hmm, there are a lot of lobster's in the ocean.  I'll bring the
> > clarified butter.
> 
> Hey, wait!!  I want to eat fish raw!!

Hahaha...  American imperialism wins again!

	- Dan C.

Re: [9fans] OT: ZFS

[YAMANASHI Takeshi]

On Thu Sep 16 09:13:58 JST 2004, boyd, rounin wrote:
> lobster sashimi is quite special.

It would get more expensive after
the Sun had boilded the ocean.. *sigh*.

By the way, the ZFS story claims that ZFS as
"the world's first 128-bit file system".

That's fine.  Ok.
But the world's first 160-bit file system is venti, isn't it?
-- 

Re: [9fans] OT: ZFS

[Russ Cox]

> That's fine.  Ok.
> But the world's first 160-bit file system is venti, isn't it?

venti is a more of a strange disk than a file system.
vac (and therefore fossil) use 56-bit file sizes.
all the disk offsets inside the venti implementation
are 64 bits.

russ

Re: [9fans] OT: ZFS

[bmaroshe]

It seems very unlikely to me, that Europe will not follow this path in a few weeks.

"We want to build picket fences around the technologies that we think are most important for the future."
 --AT&T IP department vice president Jeff George in 2000

"You get value from patents in two ways: through fees, and through licensing negotiations that give IBM access to other patents. The IBM patent portfolio gains us the freedom to do what we need to do through cross-licensing--it gives us access to the inventions of others that are the key to rapid innovation. Access is far more valuable to IBM than the fees it receives from its 9,000 active patents. There's no direct calculation of this value, but it's many times larger than the fee income, perhaps an order of magnitude larger."
 --IBM assistant general counsel Roger Smith in 1990

"If I have seen further it is by standing on the shoulders of giants."
 --Isaac Newton

boris

----- Original Message -----
From: "Ronald G. Minnich" <rminnich@lanl.gov>
Date: Friday, September 17, 2004 6:21 pm
Subject: Re: [9fans] OT: ZFS

> 
> 
> On Fri, 17 Sep 2004 rog@vitanuova.com wrote:
> 
> > could they really patent such a trivial thing
> 
> Hey, it's the US. The old saying used to be that a clever 
> prosecutor could 
> get a grand jury to indict a ham sandwich. Now you can patent it.
> 
> ron
> 

Re: [9fans] OT: ZFS

[Aharon Robbins]

> > Someone has to swap them, if they don't use the same order.
>
> Not really..  Could be ascii.  xml even.  ;-)

If you extend ASCII to directory entries, and outlaw \n in filenames,
using it instead to terminate the directory entry, you could have
cat'able directories.  That'd be pretty cool. :-)

Arnold