Linked by Thom Holwerda on Mon 15th Nov 2010 23:37 UTC, submitted by comay
Oracle and SUN Today Oracle released its latest version of Solaris technology, the Oracle Solaris 11 Express 2010.11 release. It includes a large number of new features not found in either Oracle Solaris 10 or previous OpenSolaris releases including ZFS encryption and deduplication, network-based packaging and provisioning systems, network virtualization, optimized I/O for NUMA platforms and optimized platform support including support for Intel's latest Nehalem and SPARC T3. In addition, Oracle Solaris 10 support is available from within a container/zone so migration of existing systems is greatly simplified. The release is available under a variety of licenses including a supported commercial license on a wide variety of x86 and SPARC platforms.
Order by: Score:
Intrest ?
by MacMan on Tue 16th Nov 2010 00:02 UTC
MacMan
Member since:
2006-11-19

Solar IMO is the best engineered operating system ever developed. For some reason, it never caught on like Linux, perhaps because by the time it was open sources, Linux had gained considerable momentum.

I just wonder how much interest Solaris will get now that Oracle owns it, and does seem to be becoming much more closed than Sun.

Reply Score: 6

RE: Intrest ?
by orestes on Tue 16th Nov 2010 00:27 UTC in reply to "Intrest ?"
orestes Member since:
2005-07-06

Why didn't it catch on when it was open sourced? Mostly Licensing, Sun's own seeming total lack of coherent direction, and the fact it was sort of a headache to deal with for people used to Linux based systems.

Reply Score: 4

RE[2]: Intrest ?
by flanque on Wed 17th Nov 2010 03:19 UTC in reply to "RE: Intrest ?"
flanque Member since:
2005-12-15

and the fact it was sort of a headache to deal with for people used to Linux based system

Bit of a loose reason say something like that. I'm a convert from Linux to Solaris and Solaris is far easier to manage and support in my view.

Reply Score: 2

RE[3]: Intrest ?
by orestes on Thu 18th Nov 2010 03:30 UTC in reply to "RE[2]: Intrest ?"
orestes Member since:
2005-07-06

Now it is. Back when it was first open sourced it was the bad old days of manual dependency resolution

Reply Score: 2

RE: Intrest ?
by nt_jerkface on Tue 16th Nov 2010 17:55 UTC in reply to "Intrest ?"
nt_jerkface Member since:
2009-08-26

Solaris IMO is the best engineered operating system ever developed. For some reason, it never caught on like Linux


That's because McNealy snubbed x86 servers until it was too late.

Arrogance is really what doomed Sun.

Reply Score: 3

Unlike Sun
by whartung on Tue 16th Nov 2010 01:11 UTC
whartung
Member since:
2005-07-06

This version can not be used in production, only for development. This is similar to how they license the Database and other things.

If you want to develop stuff, you can download and test and work all you want. But as soon as you deploy it, you need to get a license.

License for Solaris appears to be $1000/Socket/Year for machines with 1-4 sockets. This is for machines that are not Oracle hardware. No idea how much is costs to load it up on an older Sun box (or perhaps none of that is considered Oracle Hardware).

This pricing is not completely out of line with something like RHEL.

The next question is whether the Open Source Dump will enable anyone to put together a similar offering outside of Oracle (like CentOS for RHEL, though not identical). Not clear how feature complete the CDDL based OSS code dump will be vs Solaris Express.

Reply Score: 5

RE: Unlike Sun
by brynet on Tue 16th Nov 2010 01:13 UTC in reply to "Unlike Sun"
brynet Member since:
2010-03-02

It hasn't even been confirmed that Oracle will be releasing any source dumps.. besides some leaked memo from a few months back.

Reply Score: 2

RE: Unlike Sun
by Kebabbert on Tue 16th Nov 2010 11:56 UTC in reply to "Unlike Sun"
Kebabbert Member since:
2007-07-27

This version can be used for EVALUATION and/or DEVELOPMENT. There is no time limit specified for evaluation nor development. You could evaluate and develop some shell scripts for say, 5 years. As long as you dont violate the license you are free.

Basically, Oracle executives say that you can use it at home. But if you use it in production you need to pay. You can buy support and patches for Solaris 11 Express.

Regarding Linux, well, for instance, RedHat6 is more expensive than Solaris 11. I trust more on an mature Enterprise Unix such as Solaris, than Linux. I would rather pay support for Solaris 11 than RedHat.

"If it has to run, it runs on Solaris"

Reply Score: 4

RE[2]: Unlike Sun
by segedunum on Tue 16th Nov 2010 18:48 UTC in reply to "RE: Unlike Sun"
segedunum Member since:
2005-07-06

I trust more on an mature Enterprise Unix such as Solaris, than Linux. I would rather pay support for Solaris 11 than RedHat.

That's absolutely fabulous for you, but unfortunately there are a sizeable and significant number of people over the past ten years who don't and haven't agreed with you. Sun wheeled out the 'mature enterprise Unix' line many times and it didn't make any difference. That's why it's Oracle having a go now and not Sun.

Reply Score: 4

RE[3]: Unlike Sun
by Kebabbert on Wed 17th Nov 2010 10:25 UTC in reply to "RE[2]: Unlike Sun"
Kebabbert Member since:
2007-07-27

"I trust more on an mature Enterprise Unix such as Solaris, than Linux. I would rather pay support for Solaris 11 than RedHat.

That's absolutely fabulous for you, but unfortunately there are a sizeable and significant number of people over the past ten years who don't and haven't agreed with you. Sun wheeled out the 'mature enterprise Unix' line many times and it didn't make any difference. That's why it's Oracle having a go now and not Sun.
"
That is dumb argumentation you are using.

You are implying that Linux is better than Solaris, because more people are using Linux. Well, I guess you have just proved that Windows is better than Linux - because more people are using Windows.




Regarding Linux vs Solaris. Every serious sysadmin knows that Linux have severe problems with stability, scalability and what not. You want to see some links?


http://kerneltrap.org/Linux/Active_Merge_Windows
"The [linux source code] tree breaks every day, and it's becomming an extremely non-fun environment to work in.

We need to slow down the merging, we need to review things more, we need people to test their f--king changes!"




Linus Torvalds says Linux is bloated and huge:
http://www.theregister.co.uk/2009/09/22/linus_torvalds_linux_bloate...

"Citing an internal INTEL corp study that tracked kernel releases, Bottomley said Linux performance had dropped about two per centage points at every release, for a cumulative drop of about 12 per cent over the last ten releases. "Is this a problem?" he asked.

"We're getting bloated and huge. Yes, it's a problem," said Torvalds."



As Linux kernel Developer Andrew Morton says:

http://lwn.net/Articles/285088/
"I used to think [code quality] was in decline, and I think that I might think that it still is. I see so many regressions which we never fix.
...
it would help if people's patches were less buggy."



I have lots of links. You want to see them? I also have links that shows that Linux sucks as a file server.
http://www.enterprisestorageforum.com/sans/features/article.php/374...

"Go mkfs a 500 TB ext-3/4 or other Linux file system, fill it up with multiple streams of data, add/remove files for a few months with, say, 20 GB/sec of bandwidth from a single large SMP server and crash the system and fsck it and tell me how long it takes. Does the I/O performance stay consistent during that few months of adding and removing files? Does the file system perform well with 1 million files in a single directory and 100 million files in the file system?

My guess is the exercise would prove my point: Linux file systems have scaling issues that need to be addressed before 100 TB environments become commonplace. Addressing them now without rancor just might make Linux everything its proponents have hoped for."



Linux has scaling problems. Sure, Linux runs on super computers on Top500 (which are just a fast network with a bunch of PCs) or on a 1024 core machine from SGI Altix (which is just some blades on a fast switch) - but that is not the same thing as a running a large machine. Linux always runs on networks. Not on a single large computer.

Reply Score: 2

RE[4]: Unlike Sun
by segedunum on Sat 20th Nov 2010 23:27 UTC in reply to "RE[3]: Unlike Sun"
segedunum Member since:
2005-07-06

Regarding Linux vs Solaris. Every serious sysadmin knows that Linux have severe problems with stability, scalability and what not.

I am a sysadmin sweetheart, as are the counltes people who dumped Solaris for Linux over the past ten years, and they didn't come back to Solaris as Sun thought they would. Sun totted out those stability and scalability problems that people weren't having many times and it didn't stop them from ending up losing 100 million a month. Sun is over, as is Solaris if Oracle doesn't do something.

Don't worry, I do know some people like you who are steadfast to the last and it's very sad to watch.

Edited 2010-11-20 23:28 UTC

Reply Score: 2

RE[2]: Unlike Sun
by dukes on Tue 16th Nov 2010 20:01 UTC in reply to "RE: Unlike Sun"
dukes Member since:
2005-07-06

This version can be used for EVALUATION and/or DEVELOPMENT. There is no time limit specified for evaluation nor development. You could evaluate and develop some shell scripts for say, 5 years. As long as you dont violate the license you are free.

Basically, Oracle executives say that you can use it at home. But if you use it in production you need to pay. You can buy support and patches for Solaris 11 Express.

Regarding Linux, well, for instance, RedHat6 is more expensive than Solaris 11. I trust more on an mature Enterprise Unix such as Solaris, than Linux. I would rather pay support for Solaris 11 than RedHat.

"If it has to run, it runs on Solaris"



Actually you can install into production as well.

#11 of the faq says:

Licensing and Support for Oracle Solaris 11 Express

11-Can I get support for Oracle Solaris 11 Express?

Yes. Oracle Solaris 11 Express is covered under the Oracle Premier Support for Operating Systems or Oracle Premier Support for Systems support option for Oracle hardware, and Oracle Solaris Premier Subscription for non-Oracle hardware. Customers must choose either of these support options should they wish to deploy Oracle Solaris 11 Express into a production environment.

http://bit.ly/cz3hgP

Reply Score: 1

nice!
by poundsmack on Tue 16th Nov 2010 02:09 UTC
poundsmack
Member since:
2005-07-13

really looking forward to testing this out tomorrow. I am hoping I can go back to using Solaris Express as my desktop OS.

Reply Score: 3

gsyoungblood
Member since:
2007-01-09

I was also excited by the Solaris 11 Express release. So much so I downloaded it before reading the license. Basically you can play with it, but you are not allowed to actually USE it. As a result, I deleted the ISO image I downloaded. Pity too, because I was really looking forward to playing with it.

My laptop is currently running OpenIndiana b147. I was going to install this on it, but now I'm not. My OpenSolaris servers have all been upgraded to OpenIndiana as well.

For those expressing interest and/or admiration in Solaris, but not willing to take on Oracle's restrictive licensing or risk licensing issues, take a look at OpenIndiana.

Reply Score: 2

brynet Member since:
2010-03-02

Yeah, licensing situation is reminds me of QNX, sad really.

Reply Score: 2

Luminair Member since:
2007-03-30

I hear FreeBSD is nice this time of year

Reply Score: 8

libray Member since:
2005-08-27

I ran OpenIndiana, but I think if your servers are Sun hardware or you're running a personal desktop, you'd be better off running Solaris 11 Express. I upgraded my Opensolaris b134 to Open Indiana b147 last month, and last night upgraded that to Solaris 11 Express.

Open Indiana is still working on releasing b148 while this Express release is b151a.

http://wiki.openindiana.org/oi/oi_148

The relationship between Oracle and Illumos is non-existant. Illumos won't be able to help you with problems on a production server. Just look at their mailing list.

Reply Score: 2

ZFS Crypto
by Dubhthach on Tue 16th Nov 2010 08:30 UTC
Dubhthach
Member since:
2006-01-12

I see ZFS crypto is part of the release. Oracle very cleverly held off integration until the build after they pull down the firewall curtain (b148). The code for ZFS crypto as a result won't come out until the full Solaris 11 release is out (2011). Stops people like Nexenta from competing with Oracle/Sun storage boxes.

Reply Score: 3

RE: ZFS Crypto
by evert on Tue 16th Nov 2010 09:38 UTC in reply to "ZFS Crypto"
evert Member since:
2005-07-06

Agreed. And for me, I really want be able to encrypt a disk partition. ZFS crypto is a /big/ feature.

Reply Score: 2

RE[2]: ZFS Crypto
by Dubhthach on Tue 16th Nov 2010 10:46 UTC in reply to "RE: ZFS Crypto"
Dubhthach Member since:
2006-01-12

Opensolaris had gotten support for the onboard AES encryption in recent Intel chips (Nehalem die-shrink: Westmere) so if you have a laptop with one of the newer i5/i7 you'll get better performance for ZFS crypto if using AES.

Darren Moffat has two good blog posts up about ZFS-Crypto (he was project lead). Obviously posted them to tie in with this release.
http://blogs.sun.com/darren/

Reply Score: 1

RE[2]: ZFS Crypto
by phoenix on Wed 17th Nov 2010 22:12 UTC in reply to "RE: ZFS Crypto"
phoenix Member since:
2005-07-11

Agreed. And for me, I really want be able to encrypt a disk partition. ZFS crypto is a /big/ feature.


ZFS crypto will not help with encrypting a single disk partition, as it works at the filesystem layer, not the disk layer. Unless you are going to make a pool using just that one partition, but then you lose all the benefits of running ZFS as you have 0 redundancy.

There are other options to encryption, even with ZFS; at least in the FreeBSD world, thanks to GEOM. Use geli(4) to create encrypted GEOM providers, then build your ZFS pool out of those providers. Voila! Encrypted ZFS pool. Sure, it's pool-level and not ZFS filesystem level, but it's still encrypted. ;)

Reply Score: 2

RE[3]: ZFS Crypto
by Kebabbert on Thu 18th Nov 2010 10:43 UTC in reply to "RE[2]: ZFS Crypto"
Kebabbert Member since:
2007-07-27

In ZFS it is extremely easy and quick to a new filesystem. Filesystems are extremely light weight, like an ordinary directory.
# zfs create myZPOOL/thisIsAFilesystem
which takes less than one second.

In other words, it is no problem that "only" filesystems are encrypted. You should only store data in filesystems anyway.

Reply Score: 2

ZFS
by Laurence on Tue 16th Nov 2010 11:28 UTC
Laurence
Member since:
2007-03-26

I've become somewhat of a ZFS fanboy in recent years but got turned off OpenSolaris when Oracle took over and the whole hoopla kicked off about the state of OpenSolaris.
Now it's looking ever more likely that ZFS will be binned as Oracle already have a vested interest in BtrFS, Linux (due to CDDL vs GPL licensing) favours BtrFS and even Apple have dropped their implementation in OS X.

It seems (aside *Solaris), FreeBSD stands alone in it's support for ZFS, and that worries me a lot.

I'm sure BtrFS excels in many ways but ZFS has already been adopted on my home file server, I've grown comfortable with it and it's (for me at least) proven itself. So the last thing I want to do is migrate to another platform.

tl;dr version: I'm really worried about the future of ZFS.

Reply Score: 3

RE: ZFS
by Kebabbert on Tue 16th Nov 2010 11:51 UTC in reply to "ZFS"
Kebabbert Member since:
2007-07-27

I dont get it. What is it to worry about the future of ZFS?


I mean, BTRFS is a joke. Sure, it will have lots of cool features (copying every ZFS function) but that is far away in the future. As of now, BTRFS is unstable and corrupts data. Have you not read the BTRFS mail lists? I would never ever trust my data on BTRFS, maybe in 10 years. But not now.

And, why should Oracle devote resources to make BTRFS better than ZFS? It will take many years before BTRFS catches up on where ZFS is today. But in 10 years, ZFS will have got much better than today. In short, BTRFS will never catch up.

Oracle earns money on ZFS today, they sell ZFS products today. Why would Oracle can ZFS, and wait 5 years until BTRFS catches up on where ZFS is today, and then wait another 5 years before BTRFS have surpasses ZFS - in total 10 years before Oracle can earn money?

From a business perspective, if you have two similar products - one mature and superior and already earns money, and one in alpha stage and buggy and only consumes money - you should kill of the weaker product. It is a simple as that.

No, the most logical step, from a pure business perspective, would be to kill BTRFS - dont you agree? Actually,
"I'm really worried about the future of BTRFS".

I really dont understand your view point.

Reply Score: 5

RE[2]: ZFS
by Laurence on Tue 16th Nov 2010 12:53 UTC in reply to "RE: ZFS"
Laurence Member since:
2007-03-26

The point being if only Solaris is supporting ZFS (and taking into account Oracle doesn't favour open source like Sun did), then ZFS might quickly become obscure tech only available to Oracle licensees.

As BtrFS is GPL, I could see many people favouring BtrFS for the long game than an increasingly closed ZFS for the short game. In fact we already see this with Linux (though Linuxes situation differs slightly due to license incomparability so improving BtrFS might prove quicker than reverse engineering ZFS for the sake of having GPL code), but thankfully I'm not loyal to Linux like many of my fellow open source advocates are. But lets face it, as much as I like Unix, Linux does have massive clout these days so there is already a massive demand for BtrFS.

So if 5 years down the line, the majority of the industry outside of Oracle are favouring BtrFS, then why would Oracle care about ZFS internally? Particularly when I'm yet to see any evidence that they're migrating their own databases from Redhat to Solaris (or at least they're not on any of the servers my organisation leases from them)

There has been plenty of occasions where the more mature of two technologies have fallen out of favour. So I wouldn't say ZFS was sitting comfortably just because it has a few years on it's rivals. If anything, I would say it's fucture was uncertain because FreeBSD is the only non-Oracle OS that seems to properly support that file system.

Edited 2010-11-16 13:10 UTC

Reply Score: 2

RE[3]: ZFS
by jessesmith on Tue 16th Nov 2010 13:30 UTC in reply to "RE[2]: ZFS"
jessesmith Member since:
2010-03-11

That doesn't make any sense at all, ZFS is open source. It's available on FreeBSD and there is a module in the works for Linux. Oracle could stop developing it today and ZFS would continue to be improved in OpenIndiana, FreeBSD and (probably) Linux.

Reply Score: 1

RE[4]: ZFS
by Laurence on Tue 16th Nov 2010 14:41 UTC in reply to "RE[3]: ZFS"
Laurence Member since:
2007-03-26

That doesn't make any sense at all, ZFS is open source. It's available on FreeBSD and there is a module in the works for Linux. Oracle could stop developing it today and ZFS would continue to be improved in OpenIndiana, FreeBSD and (probably) Linux.

But my point is unless ZFS's Linux module is GPL (ie a complete re-write) then Linux is unlikely to adopt it as standard over BtrFS.

So should Oracle drop ZFS (hypothetical question i know) will there be enough community developers to keep ZFS alive?

Maybe I'm worrying over nothing. I very much hope that's the case, but I guess only time will tell

Reply Score: 2

RE[3]: ZFS
by Kebabbert on Tue 16th Nov 2010 14:26 UTC in reply to "RE[2]: ZFS"
Kebabbert Member since:
2007-07-27

I dont agree with your post. I suggest you stop worrying about ZFS and start to worry about BTRFS instead.

One of the great things about ZFS is because your data is safe with ZFS. I doubt your data is safe with BTRFS, as your data is not safe with XFS, JFS, ReiserFS, hardware raid-5 or 6, etc - which is showed in current research. I therefore doubt BTRFS keeps your data safe.

Why do you think ZFS is slower than BTRFS? Because ZFS does those checksums to guarantee your data is safe.

People that prioritize data safety, will surely continue to use ZFS. BTRFS is just another Linux filesystem - and as we all know, Linux filesystems are not really ready for Enterprise use, as they scale bad. You want to see links and articles about this?

Reply Score: 3

RE[4]: ZFS
by Laurence on Tue 16th Nov 2010 14:48 UTC in reply to "RE[3]: ZFS"
Laurence Member since:
2007-03-26

I dont agree with your post. I suggest you stop worrying about ZFS and start to worry about BTRFS instead. One of the great things about ZFS is because your data is safe with ZFS. I doubt your data is safe with BTRFS, as your data is not safe with XFS, JFS, ReiserFS, hardware raid-5 or 6, etc - which is showed in current research. I therefore doubt BTRFS keeps your data safe.

You've missed my point. I'm not worried about my data (as I said when I opened this discussion, I'm somewhat of a ZFS fanboy). I'm worried about the future development of ZFS.


Why do you think ZFS is slower than BTRFS?

I never said nor even suggested it was slower.

BTRFS is just another Linux filesystem - and as we all know, Linux filesystems are not really ready for Enterprise use, as they scale bad.

BtrFS isn't just "another Linux file system" though. It's built by Oracle and Oracle are all about salability and enterprise use. BtrFS is also the underlying file system behind Oracles distributed file system (the name of which escapes me). So Oracle do have a vested interest in making BtrFS work as much as they do with ZFS.

You want to see links and articles about this?

Actually yes please. Not because I dispute your claims but more because I'm a geek and enjoy this kind of reading material ;)

Reply Score: 3

RE[5]: ZFS
by Kebabbert on Wed 17th Nov 2010 10:29 UTC in reply to "RE[4]: ZFS"
Kebabbert Member since:
2007-07-27

"You want to see links and articles about this?

Actually yes please. Not because I dispute your claims but more because I'm a geek and enjoy this kind of reading material ;)
"
See one of my comments on Linux shortcomings on page nr 4. It is the 48th comment or so.

Reply Score: 2

RE[5]: ZFS
by phoenix on Wed 17th Nov 2010 22:52 UTC in reply to "RE[4]: ZFS"
phoenix Member since:
2005-07-11

"BTRFS is just another Linux filesystem - and as we all know, Linux filesystems are not really ready for Enterprise use, as they scale bad.

BtrFS isn't just "another Linux file system" though.
"

Actually, it is "just another filesystem", one amongst many. No distribution is using it as the default.

It's built by Oracle and Oracle are all about scalability and enterprise use.


Yes, but btrfs code in the kernel had (until very very very recently) the EXPERIMENTAL marker still on it. The on-disk format may be stable, but the fs itself most certainly is not. And the userland tools are changing on an almost daily basis, something that should not happen in an "enterprise-ready fs".

Oracle may be about scalability, but btrfs most certainly is not. Not yet, and not for at least another year or two. They don't even have parity RAID support yet.

So Oracle do have a vested interest in making BtrFS work as much as they do with ZFS.


Oracle wants to make money. Selling hardware makes them lots of money. What was one of Sun's better selling line of harwdare? Storage systems using ZFS. There's no way they're going to give up that cash cow. Btrfs won't run on the kinds of storage servers that ZFS will.

You're making a mountain out of an ant hill. ZFS isn't going anywhere for at least 5 years. Btrfs won't be usable for at least 5 years. IOW, you have 5 years of peace ahead of you before you need to start worrying about things.

Reply Score: 2

RE[2]: ZFS
by nt_jerkface on Tue 16th Nov 2010 17:45 UTC in reply to "RE: ZFS"
nt_jerkface Member since:
2009-08-26


No, the most logical step, from a pure business perspective, would be to kill BTRFS - dont you agree? Actually,
"I'm really worried about the future of BTRFS".


That does make business sense given they are investing in Solaris to make it more competitive with Linux.

Why should they also fund the development of tech that improves Linux? Funding BTRFS made more sense before the purchase of Sun.

Reply Score: 2

RE[3]: ZFS
by segedunum on Tue 16th Nov 2010 19:03 UTC in reply to "RE[2]: ZFS"
segedunum Member since:
2005-07-06

That does make business sense given they are investing in Solaris to make it more competitive with Linux.

So did Sun.........

Reply Score: 2

RE[4]: ZFS
by nt_jerkface on Tue 16th Nov 2010 19:12 UTC in reply to "RE[3]: ZFS"
nt_jerkface Member since:
2009-08-26

"That does make business sense given they are investing in Solaris to make it more competitive with Linux.

So did Sun.........
"

Sun didn't care about business sense when Schwartz was in charge. He did after all state that they were not working with a business model.

Reply Score: 3

RE[3]: ZFS
by sorpigal on Tue 16th Nov 2010 19:08 UTC in reply to "RE[2]: ZFS"
sorpigal Member since:
2005-11-02

What would make the most sense, now, is to shoehorn ZFS into Linux the way Novell put NSS into Linux (ie, in a way which sucks and is largely impossible to replicate outside their own distro) and then drop btrfs like a hot-potato. "Want ZFS Linux? Only sold by Oracle, baby!" I can see it now.

Reply Score: 3

RE: ZFS
by Dubhthach on Tue 16th Nov 2010 12:04 UTC in reply to "ZFS"
Dubhthach Member since:
2006-01-12

... and even Apple have dropped their implementation in OS X.


ZFS was originaly touted on Apple website for one of previous releases of OSX (10.5 or 10.6 can't remember) Of course it didn't help that Jonathan Schwartz had to open his big mouth and announce that it was gonna be default filesystem in OSX. Jobs just doesn't like other people pre-announcing stuff. Given the rapprochement between Oracle/Apple over Java we might be lucky and see ZFS as part of 10.7 (lion).

That's more of a pipedream on my part but you never know. The recent "back to Mac" event said very little about Lion other then about some of the GUI stuff and the release name. Given that OSX currently uses DTrace and Solaris NFSv4 stack (client/server) I wouldn't be surprised if ZFS snuck back in. Helps that the NetApp case has ended. Plus ZFS is even better now then it was two years ago with stuff like DeDup and Crypto now included.

Reply Score: 2

RE[2]: ZFS
by MacMan on Tue 16th Nov 2010 13:13 UTC in reply to "RE: ZFS"
MacMan Member since:
2006-11-19

Given the rapprochement between Oracle/Apple over Java we might be lucky and see ZFS as part of 10.7 (lion).


That would be freaking nice!

The other features from Solaris I'd like to see for Lion are:
1: Solaris memory manager
2: Solaris scheduler
3: Solaris network stack
4: Solaris process management
5: Solaris init system

Eh, screw it, I'd like to see everything beneath the window server replaced with Solaris - now that would rock.

Reply Score: 3

RE[2]: ZFS
by kaiwai on Tue 16th Nov 2010 13:26 UTC in reply to "RE: ZFS"
kaiwai Member since:
2005-07-06

ZFS was originaly touted on Apple website for one of previous releases of OSX (10.5 or 10.6 can't remember) Of course it didn't help that Jonathan Schwartz had to open his big mouth and announce that it was gonna be default filesystem in OSX. Jobs just doesn't like other people pre-announcing stuff. Given the rapprochement between Oracle/Apple over Java we might be lucky and see ZFS as part of 10.7 (lion).

That's more of a pipedream on my part but you never know. The recent "back to Mac" event said very little about Lion other then about some of the GUI stuff and the release name. Given that OSX currently uses DTrace and Solaris NFSv4 stack (client/server) I wouldn't be surprised if ZFS snuck back in. Helps that the NetApp case has ended. Plus ZFS is even better now then it was two years ago with stuff like DeDup and Crypto now included.


From what I understood it all came down to patents and licensing plus other stuff - stuff that could have over come if Sun (and subsequently Oracle) had the will power to do so. Even if they did get ZFS working with Mac OS X the problem I found when using it with OpenSolaris was that the performance was terrible on anything less than 2GB of RAM using a 32bit kernel on a low powered machine - which would have pretty much killed off using it on the MacBook Air. What Apple needs isn't something revolutionary but rather something that is evolutionary and does what is required without too much fanfare.

One possible replacement for HFS+ that comes to mind could be HAMMERFS from DragonflyBSD:

http://en.wikipedia.org/wiki/HAMMER

It includes many cool features and is licensed under a liberal license which will allow Apple to bring it over to XNU and make the necessary enhancements for Finder integration. Sure, it isn't a massive leap in the case of ZFS but the file system sticks to the tried and true way of doing things - I'd sooner have that than risk a jump into the unknown for the sake of having something on the bleeding edge.

Reply Score: 2

RE[3]: ZFS
by Kebabbert on Tue 16th Nov 2010 14:22 UTC in reply to "RE[2]: ZFS"
Kebabbert Member since:
2007-07-27

the problem I found when using ZFS with OpenSolaris was that the performance was terrible on anything less than 2GB of RAM using a 32bit kernel on a low powered machine

The problem was the 32bit cpu. ZFS is 128 bit, and does not like 32 bit cpus, performance will suck. I used 32bit Pentium4 and 1GB RAM for a year, and I got 20-30MB/sec with a ZFS raid with 4 discs. This sucks indeed.

If you use 64 bit cpus, then performance will be normal.

The reason ZFS gets slow, is because ZFS does lots of checksumming and computations with respect to data integrity. ZFS wants to guarantee that your data is not subject to bit rot or silent corruption. Other filesystems or solutions, dont do that. Why do you think they are much faster than ZFS? Because they dont do the expensive calculations with respect to data safety!

Reply Score: 2

RE[4]: ZFS
by kaiwai on Tue 16th Nov 2010 17:30 UTC in reply to "RE[3]: ZFS"
kaiwai Member since:
2005-07-06

The problem was the 32bit cpu. ZFS is 128 bit, and does not like 32 bit cpus, performance will suck. I used 32bit Pentium4 and 1GB RAM for a year, and I got 20-30MB/sec with a ZFS raid with 4 discs. This sucks indeed.

If you use 64 bit cpus, then performance will be normal.


But on low powered 64bit processors such as the MacBook Air 1.4Ghz the experience isn't exactly going to be all that pleasant to say the least. The amount of memory is a big killer - give it 4GB of memory and it flies but I say 2GB minimum but even then it isn't all too pleasant. It is a great file system designed for a large system with a tonne of memory but for something that is low powered requiring something light weight it is probably not the ideal file system to use.

The reason ZFS gets slow, is because ZFS does lots of checksumming and computations with respect to data integrity. ZFS wants to guarantee that your data is not subject to bit rot or silent corruption. Other filesystems or solutions, dont do that. Why do you think they are much faster than ZFS? Because they dont do the expensive calculations with respect to data safety!


People must have some really bad luck because of all the things that have pulled down my computer I've never experienced file system corruption - 9/10 if something goes haywire it is because of my own doing rather than an act of the computer gods trying to smite me with a kernel panic and file system corruption.

In the 10 years I've owned Mac's (from an eMac to the current iMac and MacBook Pro) I haven't experienced a single case of file system corruption *touch wood*

Reply Score: 2

RE[5]: ZFS
by Kebabbert on Wed 17th Nov 2010 10:15 UTC in reply to "RE[4]: ZFS"
Kebabbert Member since:
2007-07-27

The amount of memory is a big killer - give it 4GB of memory and it flies but I say 2GB minimum but even then it isn't all too pleasant. It is a great file system designed for a large system with a tonne of memory but for something that is low powered requiring something light weight it is probably not the ideal file system to use.

Yes, it is officially known that ZFS is for Enterprise servers. Not desktop. Sun thinks it is safe to assume that an Enterprise server for $10.000 USD has at least 2GB RAM and 64 bit cpu?

However, the slow speed is secondary. Is your data safe? With ordinary filesystems your data is not safe. Neither with hardware raid. See below



People must have some really bad luck because of all the things that have pulled down my computer I've never experienced file system corruption
...
In the 10 years I've owned Mac's (from an eMac to the current iMac and MacBook Pro) I haven't experienced a single case of file system corruption *touch wood*

Or maybe you have not noticed the signs of silent corruption? If silent corruption happens, what will the result be? That the computer halts with a blinking sign "SILENT CORRUPTION"?

No. Maybe you can not open your file. Or your computer crashes - and the crash is blamed on an "shitty OS".



The reason you did not earlier experience silent corruption or bit rot, is because your drive was small. 30% of a drives surface is devoted to error checking and correcting codes - there are errors all the time. Most of them are corrected on the fly. A tiny percent are not correctable nor detectable. See the spec for a new Enterprise disk, it will say "1 irreparable bit on 10^15 bits" - those errors are not correctable.

If you have a tiny drive, you will never reach 10^15 bits. If you have a large array with large disks, you will quickly reach 10^15 bits. During these last 10 years, you had 1GB drives - and you never reached 10^15 bits. Today you will.

If you value your data, I suggest you read this CERN study about bit rot. CERN did at study on their storage servers and discovered lot of bit rot - even the hardware did not know!
http://storagemojo.com/2007/09/19/cerns-data-corruption-research/

Or this link where a researcher shows that XFS, JFS, NTFS, ReiserFS etc might corrupt your data:
http://www.zdnet.com/blog/storage/how-microsoft-puts-your-data-at-r...

I have lot of links about bit rot. You want to read them? This is one of the most important articles on bit rot:
http://queue.acm.org/detail.cfm?id=1317400

Because no solution (except ZFS) provides data safety, I doubt BTRFS will provide data safety.

Researchers have examined ZFS with respect to data safety, and ZFS seems to be safe. You want to read the research paper?

Reply Score: 3

RE[3]: ZFS
by Laurence on Tue 16th Nov 2010 14:57 UTC in reply to "RE[2]: ZFS"
Laurence Member since:
2007-03-26

What Apple needs isn't something revolutionary but rather something that is evolutionary and does what is required without too much fanfare. One possible replacement for HFS+ that comes to mind could be HAMMERFS from DragonflyBSD: http://en.wikipedia.org/wiki/HAMMER It includes many cool features and is licensed under a liberal license which will allow Apple to bring it over to XNU and make the necessary enhancements for Finder integration. Sure, it isn't a massive leap in the case of ZFS but the file system sticks to the tried and true way of doing things - I'd sooner have that than risk a jump into the unknown for the sake of having something on the bleeding edge.


I've not read up on HAMMERFS yet (will read your wiki shortly), but ZFS would have been perfect for the media professional market. Many OS X users are photographers, musicians, music producers or movie producers. They deal with huge files, many of which are duplicates of each other minus small alterations. So ZFSs deduping and/or the intelligent copy feature (where only the differences are stored when a file is copied).

Reply Score: 2

RE[4]: ZFS
by kaiwai on Tue 16th Nov 2010 17:36 UTC in reply to "RE[3]: ZFS"
kaiwai Member since:
2005-07-06

I've not read up on HAMMERFS yet (will read your wiki shortly), but ZFS would have been perfect for the media professional market. Many OS X users are photographers, musicians, music producers or movie producers. They deal with huge files, many of which are duplicates of each other minus small alterations. So ZFSs deduping and/or the intelligent copy feature (where only the differences are stored when a file is copied).


Actually data de-dup is one of the features of HAMMERFS ;)

http://en.wikipedia.org/wiki/HAMMER

* Keeps configurable fine-grained filesystem history (30-second boundaries typically).
* Keeps configurable coarse-grained filesytem history (60 days daily by default).
* History and snapshots are accessible via the live filesystem.
* Near real-time log-less streaming mirroring to slaves or backups.
* Slave can have different retention parameters.
* CRC checksumming of data and metadata.
* Minimal remount time (no fsck required).
* Designed from the ground up for large fan-out mirroring.
* Support for very large file-systems (up to 1 exabyte).
* Offline recovery after a catastrophic failure is possible.
* Ability to re-optimize the layout in the background, on a live filesystem.
* Data de-dup
* Future - multi-master clustering
* Future - data compression


I've given DragonflyBSD a tinker and what I notice is there isn't the same sort of performance penalty as I notice with ZFS so that alone makes me a big booster for the file system; first impressions are sure to win me over and HAMMERFS has definitely done that ;)

Reply Score: 2

RE[5]: ZFS
by fithisux on Tue 16th Nov 2010 18:00 UTC in reply to "RE[4]: ZFS"
fithisux Member since:
2006-01-22

I hope OpenIndiana/Ilumos switch to HammerFS. I also hope that Illumos abandons the OpenSolaris DDK and switch to a port of IOKIT for easier development and sharing of knowledge with OSX users. Personally I wished for a port of XNU open source parts to Illumos with filling missing bits with OpenSolaris.

Reply Score: 2

RE[6]: ZFS
by Luminair on Tue 16th Nov 2010 21:27 UTC in reply to "RE[5]: ZFS"
Luminair Member since:
2007-03-30

hammerfs is what, 5 years and millions of dollars behind zfs? so I don't think it is appropriate to compare them like that. unless you're talking fantasy land style. like fabio romance fiction.

Reply Score: 2

RE[6]: ZFS
by Kebabbert on Wed 17th Nov 2010 10:27 UTC in reply to "RE[5]: ZFS"
Kebabbert Member since:
2007-07-27

I hope OpenIndiana/Ilumos switch to HammerFS.

Are you mad? HammerFS is most probably not safe, and might corrupt your data.

Researchers show that ZFS is safe with respect to bit rot and Silent corruption.

Reply Score: 2

RE: HAMMER de-duplication
by Lazarus on Tue 16th Nov 2010 21:21 UTC in reply to "RE[4]: ZFS"
Lazarus Member since:
2005-08-10

Yikes.

The deduplication code only hit the 2.9 branch of DragonFly on the 7th of November. Not something that is advertised as a stable feature for the current version of the HAMMER FS.

http://leaf.dragonflybsd.org/mailarchive/commits/2010-11/msg00043.h...

Reply Score: 2

RE: ZFS
by whartung on Tue 16th Nov 2010 18:13 UTC in reply to "ZFS"
whartung Member since:
2005-07-06

Now it's looking ever more likely that ZFS will be binned as Oracle already have a vested interest in BtrFS, Linux (due to CDDL vs GPL licensing) favours BtrFS and even Apple have dropped their implementation in OS X.


You do realize that if Oracle wanted to include ZFS in to Linux, they can GPL the whole kit tomorrow. They could stop funding/working on BtrFS, GPL ZFS and Linux could run it merry way with ZFS.

Reply Score: 2

RE: ZFS
by Luminair on Tue 16th Nov 2010 21:19 UTC in reply to "ZFS"
Luminair Member since:
2007-03-30

ugh, backwards. oracle sponsored btrfs as a direct answer to zfs, right? because they had linux, and didn't have solaris, right?

so see anything strange about your concern for zfs? anything sticking out here? read the news lately?

Reply Score: 2

RE: ZFS
by flanque on Wed 17th Nov 2010 03:22 UTC in reply to "ZFS"
flanque Member since:
2005-12-15

I wouldn't worry. UFS has had its time and as others have implied ZFS is a very mature, well supported and integrated filesystem for Solaris.

Reply Score: 2

RE: ZFS
by phoenix on Wed 17th Nov 2010 22:41 UTC in reply to "ZFS"
phoenix Member since:
2005-07-11

Now it's looking ever more likely that ZFS will be binned as Oracle already have a vested interest in BtrFS, Linux (due to CDDL vs GPL licensing) favours BtrFS and even Apple have dropped their implementation in OS X.


BtrFS is still 5 years away from achieving feature parity with ZFSv15 (latest available in FreeBSD 8.x), possibly longer to reach parity with ZFSv31 (latest available from Oracle).

There's no way they'd turf ZFS and try to switch all their development resources to BtrFS.

After all, that eliminates a reason to use Solaris, which eliminates the need for all those expensive licenses, and all those sales of expensive hardware. ;)

tl;dr version: I'm really worried about the future of ZFS.


ZFS is opensource, it's not going anywhere. ZFSv28, which includes everything except crypto support, is already "in the wild". Oracle can't close it up and make it disappear. There are even experimental patches for using ZFSv28 in FreeBSD 9-CURRENT, with the goal of having it integrated into the 9.0 release in 2011.

Nexenta is available now with ZFS. It's not going anywhere.

GreenBytes uses ZFS in their storage products. They aren't going anywhere anytime soon.

There are others. It's too late to put the genie back in the bottle. And it's too long to wait for BtrFS to catch up.

There are even two separate projects underway to bring ZFS to Linux as an out-of-tree kernel module.

IOW, there's nothing to worry about.

Edited 2010-11-17 22:42 UTC

Reply Score: 2

RE[2]: ZFS
by Kebabbert on Thu 18th Nov 2010 10:44 UTC in reply to "RE: ZFS"
Kebabbert Member since:
2007-07-27

Also, ZFS gives very good data safety. Here are some problems hardware raid Enterprise had
http://www.mysqlperformanceblog.com/2010/11/16/on-the-perils-of-uni...

Reply Score: 2

Oracle aims BIG
by wigry on Tue 16th Nov 2010 12:18 UTC
wigry
Member since:
2008-10-09

I hope you understand that with quite expensive license Oracle actually tries to create a competition on mainframe market where it tries to create alternatives to IBM, HP, Fujitsu big iron. There is insane amounts of money moving in mainframe business and Solaris 11 will be targeted mainly for that market. Oracle wants to be part of the BIG GAME.

Reply Score: 1

RE: Oracle aims BIG
by Kebabbert on Tue 16th Nov 2010 14:18 UTC in reply to "Oracle aims BIG"
Kebabbert Member since:
2007-07-27

Again, Solaris 11 is CHEAPER than Linux.

Reply Score: 1

RE[2]: Oracle aims BIG
by Slambert666 on Tue 16th Nov 2010 15:37 UTC in reply to "RE: Oracle aims BIG"
Slambert666 Member since:
2008-10-30

Again, Solaris 11 is CHEAPER than Linux.


Or to say it more correctly:
The cheapest Solaris 11 license is cheaper than the most expensive Linux license.

Reply Score: 4

RE[3]: Oracle aims BIG
by Kebabbert on Wed 17th Nov 2010 10:30 UTC in reply to "RE[2]: Oracle aims BIG"
Kebabbert Member since:
2007-07-27

"Again, Solaris 11 is CHEAPER than Linux.


Or to say it more correctly:
The cheapest Solaris 11 license is cheaper than the most expensive Linux license.
"
Read here about Solaris vs RedHat pricing
http://www.c0t0d0s0.org/archives/7023-New-RedHat-pricing.html

And you will see how expensive Linux can be, and Solaris is cheaper.

Reply Score: 2

Comment by kaiwai
by kaiwai on Tue 16th Nov 2010 13:06 UTC
kaiwai
Member since:
2005-07-06

I had a look through some of the changes and the biggest disappointment I see is they haven't replaced HAL with something that really has a long term development path - I only hope that maybe 'Solaris Express' is a sneak preview rather than a 'final product that needs testing'. If it is a partially finished product I hope one of their arms is to remove HAL and to address the issue of a tickless kernel because from what I've seen over on OpenSolaris.org mailing list the development seems to have almost ground to a halt/died the moment that Oracle bought out Sun.

The reason why I ask is that although 'big iron' is the focus I'm sure that Oracle also has to keep in mind that there are system admins and developers who use Solaris on their laptops and desktops/workstations as well as terminal appliances that make use of Solaris as a desktop within large organisations. It is a great to see that Sun technology has some major backing behind it but I hope that Oracle just isn't going to ram out a new release with some pretty minor additions.

Reply Score: 2

RE: Comment by kaiwai
by Luminair on Tue 16th Nov 2010 21:32 UTC in reply to "Comment by kaiwai"
Luminair Member since:
2007-03-30

when a company gets acquired and the project leadership changes and major workers leave and.... well I'm not sure why you'd expect anything significant to get done so soon

also, I'm not so sure you should be so sure about that last part

Reply Score: 2

c0t0d0s0
Member since:
2008-10-16

I read this "Hey, but Linux or BSD has the xyz filesystem" again and again. However i just want to note, that a filesystem needs several years of development to be feature complete and then several years of matureing. ZFS was started in 2001 and we see major uptake of the filesystem since 2009 or so. I just don't get it, why the people touting the zfs alikes think that those filesystem could take shortcuts on that road. There is no room for shortcuts at storing data.

The mistake many proponents of other ZFS alike filesystems do is to take the uptake time of ext(x) and think you will see the same with all this new filesystems. But they tend to forget that all the +x at ext were just incremental steps. Nothing really new.

We will see a lot of end-of-year fireworks before we see uptake in areas were it really matters. Before that it's a nice toy.

Reply Score: 1