[zfs-discuss] Re: Poor performance on NFS-exported ZFS volumes
Patrick Bachmann: Hey Bill, Bill Sommerfeld wrote: Overly wide raidz groups seems to be an unfenced hole that people new to ZFS fall into on a regular basis. The man page warns against this but that doesn't seem to be sufficient. Given that zfs has relatively few such traps, perhaps large raidz groups ought to be implicitly split up absent a "Yes, I want to be stupid" flag.. IMHO it is sufficient to just document this best-practice. I disagree. The documentation has to AT LEAST state that more than 9 disks gives poor performance. I did read that raidz should use 3-9 disks in the docs but it doesn't say WHY, so of course I went ahead and used 12 disks. When I say I disagree, I mean this has to be documented in the standard docs (man pages) not some best-practices guide on some wiki. But really I disagree that this needs documentation. So much of zfs is meant to be automatic, now we're back to worrying about stripe width? (Or maybe that's not the problem but it sure is the same type of manual administration.) I may have 12 disks and it simply does not make sense (for my theoretical configuration) to split them up into two pools. I would then have to worry about sizing each pool correctly. zfs is supposed to fix that problem. -frank ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] iscsi automatic mounting
zfs automatically mounts locally attached disks (export/import aside). Does it do this for iscsi? I guess my question is, does the solaris iscsi initiator provide the same kind of device permanence as for local drives? thanks -frank ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: Poor performance on NFS-exported ZFS volumes
Frank Cusack wrote: Patrick Bachmann: Hey Bill, Bill Sommerfeld wrote: Overly wide raidz groups seems to be an unfenced hole that people new to ZFS fall into on a regular basis. The man page warns against this but that doesn't seem to be sufficient. Given that zfs has relatively few such traps, perhaps large raidz groups ought to be implicitly split up absent a "Yes, I want to be stupid" flag.. IMHO it is sufficient to just document this best-practice. I disagree. The documentation has to AT LEAST state that more than 9 disks gives poor performance. I did read that raidz should use 3-9 disks in the docs but it doesn't say WHY, so of course I went ahead and used 12 disks. When I say I disagree, I mean this has to be documented in the standard docs (man pages) not some best-practices guide on some wiki. But really I disagree that this needs documentation. So much of zfs is meant to be automatic, now we're back to worrying about stripe width? (Or maybe that's not the problem but it sure is the same type of manual administration.) I may have 12 disks and it simply does not make sense (for my theoretical configuration) to split them up into two pools. I would then have to worry about sizing each pool correctly. zfs is supposed to fix that problem. This may just be a matter of wording, but you wouldn't have to split it up into two pools. You could use two smaller raidz vdevs within the same pool. -- Rich. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS vs WAFL positioning
I've had people mention that WAFL does indeed support clones of snapshots. Is this a "what version of WAFL" problem? Darren ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS vs WAFL positioning
On 7/28/06, Darren Reed <[EMAIL PROTECTED]> wrote: I've had people mention that WAFL does indeed support clones of snapshots. Is this a "what version of WAFL" problem? apparently so, but it is rather new from the impression given from this site: http://www.tournament.org.il/run/index.php?/archives/67-Ontap-Simulator,-and-some-insights-about-NetApp.html "I believe that the new "clone" method is based on the WAFL built-in snapshot capabilities. Although WAFL Snapshots are supposed to be space conservatives, they require a guaranteed space on the aggregation prior to commiting the clone itself. If the aggregation is too crowded, they will fail with the error message "not enough space". If there is enough for snapshots, but not enough to guarantee a full clone, you'll get a message saying "space not guaranteed"." James Dickens uadmin.blogspot.com Darren ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] sharing a storage array
On 7/28/06, Jeff Bonwick <[EMAIL PROTECTED]> wrote: > I have a SAS array with a zfs pool on it. zfs automatically searches for > and mounts the zfs pool I've created there. I want to attach another > host to this array, but it doesn't have any provision for zones or the > like. (Like you would find in an FC array or in the switch infrastructure.) > > Will host-b not automatically mount filesystems on pools created on host-a, > and vice versa, or is this going to be a problem. Ideally, I could create > a single pool and mount some filesystems on host-a, and some on host-b, but > barring that even just being able to have 2 pools and each can be mounted > on one of the hosts would be great. > > If one host failed I want to be able to do a manual mount on the other host. Multiple hosts writing to the same pool won't work, but you could indeed have two pools, one for each host, in a dual active-passive arrangement. That is, you dual-attach the storage with host A talking to pool A and host B talking to pool B. If host A fails, it can 'zpool import -f B' to start serving up the B data. HA-ZFS (part of SunCluster 3.2) will automate this, but for now you can roll your own along these lines. Hi okay just wondering but can you define "won't work" will ZFS spot someone else writing to the disks and refuse to do any work? will it spill its guts all over the dumpdevice? or just fight each other correcting each others changes? James Dickens uadmin.blogspot.com Jeff ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re[2]: [zfs-discuss] 6424554
Hello Fred, Friday, July 28, 2006, 12:37:22 AM, you wrote: FZ> Hi Robert, FZ> The fix for 6424554 is being backported to S10 and will be available in FZ> S10U3, later this year. I know that already - I was rather asking if a patch containing the fix will be available BEFORE U3 and if yes then when? IMHO many more people I evaluating ZFS now when it's in a stable Solaris release. Any performance fixes to ZFS should be available as soon as possible 'coz that's one of the things people are looking at and once they will be disappointed it will take long time for them to try again. Anyway I will try to get IDR via support channels. -- Best regards, Robertmailto:[EMAIL PROTECTED] http://milek.blogspot.com ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re[2]: [zfs-discuss] Zones root-fs in ZFS ? (fwd)
Hello Matty, Thursday, July 27, 2006, 7:53:34 PM, you wrote: M> Are there any known issues with patching zones that are installed on a ZFS M> file system? Does smpatch and company work ok with this configuration? Right now I have such configurations and have been using smpatch without any problems so far. -- Best regards, Robertmailto:[EMAIL PROTECTED] http://milek.blogspot.com ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] 3510 JBOD ZFS vs 3510 HW RAID
Hi there Is it fair to compare the 2 solutions using Solaris 10 U2 and a commercial database (SAP SD scenario). The cache on the HW raid helps, and the CPU load is less... but the solution costs more and you _might_ not need the performance of the HW RAID. Has anybody with access to these units done a benchmark comparing the performance (and with the pricelist in hand) came to a conclusion. It's not as such about maximum performance from both, but the price/performance. If a JBOD with ZFS does 500 IO's @ $10K vs a HW RAID 700 IO's @ $20K ... then the JBOD would be a good investment when $ is a factor. (example) Thank you. This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] Re: zfs questions from Sun customer
> > * follow-up question from customer > > > Yes, using the c#t#d# disks work, but anyone using fibre-channel storage > on somethink like IBM Shark or EMC Clariion will want multiple paths to > disk using either IBMsdd, EMCpower or Solaris native MPIO. Does ZFS > work with any of these fibre channel multipathing drivers? As a side node: EMCpower does work with ZFS: # pkginfo -l EMCpower [...] VERSION: 4.5.0_b169 # zpool create test emcpower3c warning: device in use checking failed: No such device # zfs list test NAME USED AVAIL REFER MOUNTPOINT test76K 8.24G 24.5K /test ZFS does not re-label the disk though, so you have to create an EFI label through some other means. MPxIO works out-of-the-box. This message posted from opensolaris.org ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: Poor performance on NFS-exported ZFS volumes
Hey Frank, Frank Cusack wrote: Patrick Bachmann: IMHO it is sufficient to just document this best-practice. I disagree. The documentation has to AT LEAST state that more than 9 disks gives poor performance. I did read that raidz should use 3-9 disks in the docs but it doesn't say WHY, so of course I went ahead and used 12 disks. The man pages says "The recommended number [of devices] is between 3 and 9". At the time of the discussion the "ZFS Admin Guide" didn't mention this at all and the result of this discussion is, that now it says: "If you are creating a RAID-Z configuration with many disks, as an example, a RAID-Z configuration with 14 disks is better split up into a two 7-disk groupings. RAID-Z configurations with single-digit groupings of disks should perform better." When I say I disagree, I mean this has to be documented in the standard docs (man pages) not some best-practices guide on some wiki. To me "documenting a best-practice" does not imply that it is just spelled out in some random wiki or a BluePrint but rather that it is written down "somewhere" and that this "best-practice" is not just documented in the mailing-lists archives, which in some other communities seem to be the only documentation available. But really I disagree that this needs documentation. So much of zfs is meant to be automatic, now we're back to worrying about stripe width? (Or maybe that's not the problem but it sure is the same type of manual administration.) I may have 12 disks and it simply does not make sense (for my theoretical configuration) to split them up into two pools. I would then have to worry about sizing each pool correctly. zfs is supposed to fix that problem. Richard already pointed out that you should split the devices into a number of vdevs and not pools. How is ZFS going to know what gives the best performance on your systems config? There are a lot of things you know better off-hand about your system, otherwise you need to do some benchmarking, which ZFS would have to do too, if it was to give you the best performing config. Oh, and performance isn't the only objective. See the threads started by Richard Elling in the past couple of weeks on this mailinglist. If no one has done it yet, I'll file a bug against the zpool(1M) man page to get the performance concern included. But to me the "ZFS Admin Guide" seemed to be a required reading. You might want to check it out. Greetings, Patrick ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] sharing a storage array
Jeff Bonwick wrote: If one host failed I want to be able to do a manual mount on the other host. Multiple hosts writing to the same pool won't work, but you could indeed have two pools, one for each host, in a dual active-passive arrangement. That is, you dual-attach the storage with host A talking to pool A and host B talking to pool B. If host A fails, it can 'zpool import -f B' to start serving up the B data. HA-ZFS (part of SunCluster 3.2) will automate this, but for now you can roll your own along these lines. Cool. Does this method require assigning each disk to one pool or the other, or can disks be divided into partitions before pool assignment? -- -- Jeff VICTOR Sun Microsystemsjeff.victor @ sun.com OS AmbassadorSr. Technical Specialist Solaris 10 Zones FAQ:http://www.opensolaris.org/os/community/zones/faq -- ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS Boot Disk
On Thu, Jul 27, 2006 at 07:25:55PM -0700, Matthew Ahrens wrote: > On Thu, Jul 27, 2006 at 08:17:03PM -0500, Malahat Qureshi wrote: > > Is there any way to boot of from zfs disk "work around" ?? > > Yes, see > http://blogs.sun.com/roller/page/tabriz?entry=are_you_ready_to_rumble I followed those directions with snv_38 and was unsucessful, I wonder what I did wrong. Sadly it's my work desktop and I had to stop screwing around with that and actually get work done. :) I think I'll just wait until you can install directly to ZFS. Any ETA on that, btw? -brian ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: Poor performance on NFS-exported ZFS volumes
On Fri, Jul 28, 2006 at 02:14:50PM +0200, Patrick Bachmann wrote: > systems config? There are a lot of things you know better off-hand > about your system, otherwise you need to do some benchmarking, which > ZFS would have to do too, if it was to give you the best performing > config. How hard would it be to write a tool like that? Something along the lines of: zpool bench raidz disk1 disk2 ... diskN Let ZFS figure out the best way to set up your disks for you and tell you how it should be laid out (and even offer a "just do it" flag that will let it automatically create the pool depending on how it sees it best)? Yes, I know my hardware, and so perhaps I wouldn't need such a tool, but there are a lot of people out there who don't really know the best way to use their hardware for the best performance. This would be an outstanding tool for them. -brian ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] sharing a storage array
Danger Will Robinson... Jeff Victor wrote: Jeff Bonwick wrote: If one host failed I want to be able to do a manual mount on the other host. Multiple hosts writing to the same pool won't work, but you could indeed have two pools, one for each host, in a dual active-passive arrangement. That is, you dual-attach the storage with host A talking to pool A and host B talking to pool B. If host A fails, it can 'zpool import -f B' to start serving up the B data. HA-ZFS (part of SunCluster 3.2) will automate this, but for now you can roll your own along these lines. Cool. Does this method require assigning each disk to one pool or the other, or can disks be divided into partitions before pool assignment? The problem with slicing disks and sharing the slices is that you are more prone to fatal operational mistakes. For storage where isolation is enforced, SCSI reservations are often used. SCSI reservations work on a per-LUN basis, not a per-slice basis because SCSI has no concept of slices (or partitions). A safer approach is to work only at a per-LUN level for sharing disks. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: Poor performance on NFS-exported ZFS volumes
Brian Hechinger wrote: On Fri, Jul 28, 2006 at 02:14:50PM +0200, Patrick Bachmann wrote: systems config? There are a lot of things you know better off-hand about your system, otherwise you need to do some benchmarking, which ZFS would have to do too, if it was to give you the best performing config. How hard would it be to write a tool like that? Something along the lines of: zpool bench raidz disk1 disk2 ... diskN Let ZFS figure out the best way to set up your disks for you and tell you how it should be laid out (and even offer a "just do it" flag that will let it automatically create the pool depending on how it sees it best)? The problem is that there are at least 3 knobs to turn (space, RAS, and performance) and they all interact with each other. Yes, I know my hardware, and so perhaps I wouldn't need such a tool, but there are a lot of people out there who don't really know the best way to use their hardware for the best performance. This would be an outstanding tool for them. I've got an idea I'm prototyping, but it isn't ZFS-specific, so I don't expect to tie it directly to ZFS. Stay tuned... -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] sharing a storage array
Richard Elling wrote: Danger Will Robinson... Jeff Victor wrote: Jeff Bonwick wrote: Multiple hosts writing to the same pool won't work, but you could indeed have two pools, one for each host, in a dual active-passive arrangement. That is, you dual-attach the storage with host A talking to pool A and host B talking to pool B. If host A fails, it can 'zpool import -f B' to start serving up the B data. HA-ZFS (part of SunCluster 3.2) will automate this, but for now you can roll your own along these lines. Cool. Does this method require assigning each disk to one pool or the other, or can disks be divided into partitions before pool assignment? The problem with slicing disks and sharing the slices is that you are more prone to fatal operational mistakes. For storage where isolation is enforced, SCSI reservations are often used. SCSI reservations work on a per-LUN basis, not a per-slice basis because SCSI has no concept of slices (or partitions). A safer approach is to work only at a per-LUN level for sharing disks. -- richard Now that I've gone and read the zpool man page :-[ it seems that only whole disks can be exported/imported. -- -- Jeff VICTOR Sun Microsystemsjeff.victor @ sun.com OS AmbassadorSr. Technical Specialist Solaris 10 Zones FAQ:http://www.opensolaris.org/os/community/zones/faq -- ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] 3510 JBOD ZFS vs 3510 HW RAID
On Fri, 28 Jul 2006, Louwtjie Burger wrote: reformatted > Hi there > > Is it fair to compare the 2 solutions using Solaris 10 U2 and a > commercial database (SAP SD scenario). > > The cache on the HW raid helps, and the CPU load is less... but the > solution costs more and you _might_ not need the performance of the HW > RAID. > > Has anybody with access to these units done a benchmark comparing the > performance (and with the pricelist in hand) came to a conclusion. > > It's not as such about maximum performance from both, but the > price/performance. If a JBOD with ZFS does 500 IO's @ $10K vs a HW RAID > 700 IO's @ $20K ... then the JBOD would be a good investment when $ is a > factor. (example) Or stated another way, is it more beneficial to spend the $10k on increased memory for the DB server, rather than on RAID hardware. Regards, Al Hopper Logical Approach Inc, Plano, TX. [EMAIL PROTECTED] Voice: 972.379.2133 Fax: 972.379.2134 Timezone: US CDT OpenSolaris.Org Community Advisory Board (CAB) Member - Apr 2005 OpenSolaris Governing Board (OGB) Member - Feb 2006 ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
[zfs-discuss] ZFS questions
Can someone explain to me what the 'volinit' and 'volfini' options to zfs do ? It's not obvious from the source code and these options are undocumented. Thanks, John -- John Cecere Sun Microsystems 732-302-3922 / [EMAIL PROTECTED] ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS Boot Disk
Brian Hechinger wrote: On Thu, Jul 27, 2006 at 07:25:55PM -0700, Matthew Ahrens wrote: On Thu, Jul 27, 2006 at 08:17:03PM -0500, Malahat Qureshi wrote: Is there any way to boot of from zfs disk "work around" ?? Yes, see http://blogs.sun.com/roller/page/tabriz?entry=are_you_ready_to_rumble I followed those directions with snv_38 and was unsucessful, I wonder what I did wrong. Sadly it's my work desktop and I had to stop screwing around with that and actually get work done. :) I think I'll just wait until you can install directly to ZFS. Any ETA on that, btw? While the official release of zfs-boot won't be out until Update 4 at least, we're working right now on getting enough pieces available through OpenSolaris so that users can put together a boot CD/DVD/image that will directly install a system with a zfs root. I can't give an exact date, but we're pretty close. I expect it within weeks, not months. Lori ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] iscsi automatic mounting
On Fri, Jul 28, 2006 at 12:43:37AM -0700, Frank Cusack wrote: > zfs automatically mounts locally attached disks (export/import aside). Does > it do this for iscsi? I guess my question is, does the solaris iscsi > initiator provide the same kind of device permanence as for local drives? No, not currently. We are working on more tightly integrating iSCSI with ZFS. So, for example, you could do: # zfs set iscsi=on pool/vol And have it "just work". There are a lot of details to work out, but we have a good idea of what needs to be done. - Eric -- Eric Schrock, Solaris Kernel Development http://blogs.sun.com/eschrock ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS questions
On Fri, Jul 28, 2006 at 10:52:50AM -0400, John Cecere wrote: > Can someone explain to me what the 'volinit' and 'volfini' options to zfs > do ? It's not obvious from the source code and these options are > undocumented. These are unstable private interfaces which create and destroy the /dev/zvol/* links for any volumes contained within the pool. Please do not use them in any code outside of ON, as they are subject to change at any time. - Eric -- Eric Schrock, Solaris Kernel Development http://blogs.sun.com/eschrock ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: Re[2]: [zfs-discuss] Zones root-fs in ZFS ? (fwd)
On July 28, 2006 11:42:28 AM +0200 Robert Milkowski <[EMAIL PROTECTED]> wrote: Hello Matty, Thursday, July 27, 2006, 7:53:34 PM, you wrote: M> Are there any known issues with patching zones that are installed on a ZFS M> file system? Does smpatch and company work ok with this configuration? Right now I have such configurations and have been using smpatch without any problems so far. I thought I read somewhere (zones guide?) that putting the zone root fs on zfs was unsupported. -frank ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: Poor performance on NFS-exported ZFS volumes
On July 28, 2006 2:14:50 PM +0200 Patrick Bachmann <[EMAIL PROTECTED]> wrote: Richard already pointed out that you should split the devices into a number of vdevs and not pools. I missed that. I guess I also didn't know what a vdev is, guess I know even less about this "zfs thing" than I thought. :-) thanks -frank ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: Poor performance on NFS-exported ZFS volumes
On July 28, 2006 9:09:58 AM -0400 Brian Hechinger <[EMAIL PROTECTED]> wrote: On Fri, Jul 28, 2006 at 02:14:50PM +0200, Patrick Bachmann wrote: systems config? There are a lot of things you know better off-hand about your system, otherwise you need to do some benchmarking, which ZFS would have to do too, if it was to give you the best performing config. How hard would it be to write a tool like that? Something along the lines of: zpool bench raidz disk1 disk2 ... diskN bonnie++? -frank ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] iscsi automatic mounting
On Fri, Jul 28, 2006 at 09:29:42AM -0700, Eric Schrock wrote: > On Fri, Jul 28, 2006 at 12:43:37AM -0700, Frank Cusack wrote: > > zfs automatically mounts locally attached disks (export/import aside). Does > > it do this for iscsi? I guess my question is, does the solaris iscsi > > initiator provide the same kind of device permanence as for local drives? > > No, not currently. We are working on more tightly integrating iSCSI > with ZFS. So, for example, you could do: > > # zfs set iscsi=on pool/vol > > And have it "just work". There are a lot of details to work out, but we > have a good idea of what needs to be done. Although taking another look at your question, I think I answered the wrong question. If you create any kind of pool, backed by local disks, files, iSCSI targets, etc, it will work as expected. As to whether you get the same performance, my guess would be "it depends". It depends on what iSCSI target you have on the other end, and what fabric you have in between. Not to mention the speed of your local drives, for comparison. The only real way to know is to run some benchmarks of your own. - Eirc -- Eric Schrock, Solaris Kernel Development http://blogs.sun.com/eschrock ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] 3510 JBOD ZFS vs 3510 HW RAID
On July 28, 2006 3:31:51 AM -0700 Louwtjie Burger <[EMAIL PROTECTED]> wrote: Hi there Is it fair to compare the 2 solutions using Solaris 10 U2 and a commercial database (SAP SD scenario). The cache on the HW raid helps, and the CPU load is less... but the solution costs more and you _might_ not need the performance of the HW RAID. Has anybody with access to these units done a benchmark comparing the performance (and with the pricelist in hand) came to a conclusion. It's not as such about maximum performance from both, but the price/performance. If a JBOD with ZFS does 500 IO's @ $10K vs a HW RAID 700 IO's @ $20K ... then the JBOD would be a good investment when $ is a factor. (example) ISTM the cheapest array is the best for zfs. If not, isn't any benchmark going to be specific to your application? -frank ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Zones root-fs in ZFS ? (fwd)
Right now I have such configurations and have been using smpatch without any problems so far. I thought I read somewhere (zones guide?) that putting the zone root fs on zfs was unsupported. You've missed the earlier part of this thread. Yes, it's unsupported, but the question was asked "Does it work anyway?" and the short answer is "yes, it mostly does, but there are enough complications and known problems with the patching and upgrade of this configuration that it is currently unsupported." Lori ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] 3510 JBOD ZFS vs 3510 HW RAID
Frank Cusack wrote: On July 28, 2006 3:31:51 AM -0700 Louwtjie Burger <[EMAIL PROTECTED]> wrote: Hi there Is it fair to compare the 2 solutions using Solaris 10 U2 and a commercial database (SAP SD scenario). The cache on the HW raid helps, and the CPU load is less... but the solution costs more and you _might_ not need the performance of the HW RAID. Has anybody with access to these units done a benchmark comparing the performance (and with the pricelist in hand) came to a conclusion. It's not as such about maximum performance from both, but the price/performance. If a JBOD with ZFS does 500 IO's @ $10K vs a HW RAID 700 IO's @ $20K ... then the JBOD would be a good investment when $ is a factor. (example) ISTM the cheapest array is the best for zfs. If not, isn't any benchmark going to be specific to your application? Specific to the app, the amount of data, how many other hosts might be in play, etc. etc. etc. That said a 3510 with a raid controller is going to blow the door, drive brackets, and skin off a JBOD in raw performance. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: Poor performance on NFS-exported ZFS volumes
Richard Elling wrote: How hard would it be to write a tool like that? Something along the lines of: zpool bench raidz disk1 disk2 ... diskN Let ZFS figure out the best way to set up your disks for you and tell you how it should be laid out (and even offer a "just do it" flag that will let it automatically create the pool depending on how it sees it best)? The problem is that there are at least 3 knobs to turn (space, RAS, and performance) and they all interact with each other. Good point. then how about something more like zpool bench raidz favor space disk1 ... diskN zpool bench raidz favor performance disk1 .. diskN That is, tell the analyzer which knob you are most interested in. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] How to best layout our filesystems
Robert,
The patches will be available sometime late September. This may be a
week or so before s10u3 actually releases.
Thanks,
George
Robert Milkowski wrote:
Hello eric,
Thursday, July 27, 2006, 4:34:16 AM, you wrote:
ek> Robert Milkowski wrote:
Hello George,
Wednesday, July 26, 2006, 7:27:04 AM, you wrote:
GW> Additionally, I've just putback the latest feature set and bugfixes
GW> which will be part of s10u3_03. There were some additional performance
GW> fixes which may really benefit plus it will provide hot spares support.
GW> Once this build is available I would highly recommend that you guys take
GW> it for a spin (works great on Thumper).
I guess patches will be released first (or later).
Can you give actual BUG IDs especially those related to performance?
ek> For U3, these are the performance fixes:
ek> 6424554 full block re-writes need not read data in
ek> 6440499 zil should avoid txg_wait_synced() and use dmu_sync() to issue
ek> parallelIOs when fsyncing
ek> 6447377 ZFS prefetch is inconsistant
ek> 6373978 want to take lots of snapshots quickly ('zfs snapshot -r')
ek> you could perhaps include these two as well:
ek> 4034947 anon_swap_adjust() should call kmem_reap() if availrmem is low.
ek> 6416482 filebench oltp workload hangs in zfs
ok, thank you.
Do you know if patches for S10 will be released before U3?
ek> There won't be anything in U3 that isn't already in nevada...
I know that :)
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] sharing a storage array
Richard Elling wrote: Danger Will Robinson... Jeff Victor wrote: Jeff Bonwick wrote: If one host failed I want to be able to do a manual mount on the other host. Multiple hosts writing to the same pool won't work, but you could indeed have two pools, one for each host, in a dual active-passive arrangement. That is, you dual-attach the storage with host A talking to pool A and host B talking to pool B. If host A fails, it can 'zpool import -f B' to start serving up the B data. HA-ZFS (part of SunCluster 3.2) will automate this, but for now you can roll your own along these lines. Cool. Does this method require assigning each disk to one pool or the other, or can disks be divided into partitions before pool assignment? The problem with slicing disks and sharing the slices is that you are more prone to fatal operational mistakes. For storage where isolation is enforced, SCSI reservations are often used. SCSI reservations work on a per-LUN basis, not a per-slice basis because SCSI has no concept of slices (or partitions). A safer approach is to work only at a per-LUN level for sharing disks. Also from a performance point of view... If the LUN is a local disk, sharing slices can turn what would be sequential disk I/O into random I/O, not a good thing if the slices are used concurrently. If the LUN is on a storage array, then per-LUN caching, read-ahead, write-behind, RAID overhead, striped sized I/Os, are all impacted. Sharing slices on a single LUN can and does cause unforeseen performance problems, often very hard to diagnose. The old KISS policy of creating one big LUN, then carving it up into disk slices or volume manager controlled volumes, often causes problems later on in the LUN's life. Jim -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS Boot Disk
On Fri, Jul 28, 2006 at 09:47:48AM -0600, Lori Alt wrote: > > While the official release of zfs-boot won't be out > until Update 4 at least, we're working right now on > getting enough pieces available through OpenSolaris > so that users can put together a boot CD/DVD/image > that will directly install a system with a zfs > root. I can't give an exact date, but we're pretty > close. I expect it within weeks, not months. That's great Lori!! I could wait months, but I'd rather wait weeks. :) What about Express? I don't have a problem running express. In fact, the desktop at work (a Dell POS) has snv_38 on it and the work laptop (a Dell POS, see a patern here? *G*) has snv_40 on it. ;) That being said, I'm (hopefuly safely) assuming that if this makes it into Update 4 it will include support for zfs-root/install on SPARC as well as x86? That would make my U80 at home very, very happy. :) -brian ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS Boot Disk
Brian Hechinger wrote: On Fri, Jul 28, 2006 at 09:47:48AM -0600, Lori Alt wrote: While the official release of zfs-boot won't be out until Update 4 at least, we're working right now on getting enough pieces available through OpenSolaris so that users can put together a boot CD/DVD/image that will directly install a system with a zfs root. I can't give an exact date, but we're pretty close. I expect it within weeks, not months. That's great Lori!! I could wait months, but I'd rather wait weeks. :) What about Express? Probably not any time soon. If it makes U4, I think that would make it available in Express late this year. That being said, I'm (hopefuly safely) assuming that if this makes it into Update 4 it will include support for zfs-root/install on SPARC as well as x86? We expect to release SPARC support at the same time as x86. Lori ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: Poor performance on NFS-exported ZFS volumes
Joseph Mocker wrote: Richard Elling wrote: The problem is that there are at least 3 knobs to turn (space, RAS, and performance) and they all interact with each other. Good point. then how about something more like zpool bench raidz favor space disk1 ... diskN zpool bench raidz favor performance disk1 .. diskN That is, tell the analyzer which knob you are most interested in. I wish it was that easy. If I optimize for space, I'll always get a big, fat RAID-0. If I optimize for RAS, I'll get multi paned (N-way) mirror. The tool has to be able to handle the spectrum in between those extremes. -- richard ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re[2]: [zfs-discuss] sharing a storage array
Hello Jeff, Friday, July 28, 2006, 4:21:42 PM, you wrote: JV> Now that I've gone and read the zpool man page :-[ it seems that only whole JV> disks can be exported/imported. No, it's not that way. If you create a pool from slices you'll be able to import/export only those slices. So if you will create two slices on one shared LUN you will be able to import each pool on a different server. However such config could be unfortunate due to other reasons others have stated. -- Best regards, Robertmailto:[EMAIL PROTECTED] http://milek.blogspot.com ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] Re: ZFS questions (hybrid HDs)
On Jun 21, 2006, at 11:05, Anton B. Rang wrote: My guess from reading between the lines of the Samsung/Microsoft press release is that there is a mechanism for the operating system to "pin" particular blocks into the cache (e.g. to speed boot) and the rest of the cache is used for write buffering. (Using it as a read cache doesn't buy much compared to using the normal drive cache RAM for that, and might also contribute to wear, which is why read caching appears to be under OS control rather than automatic.) Actually, Microsoft has been posting a bit about this for the upcoming Vista release .. WinHEC '06 had a few interesting papers and it looks like Microsoft is going to be introducing SuperFetch, ReadyBoost, and ReadyDrive .. mentioned here: http://www.microsoft.com/whdc/system/sysperf/accelerator.mspx The ReadyDrive paper seems to outline their strategy on the industry Hybrid Drive push and the recent t13.org adoption of the ATA-ACS8 command set: http://www.microsoft.com/whdc/device/storage/hybrid.mspx It also looks like they're aiming at some sort of driver level PriorityIO scheme which should play nicely into lower level tiered hardware in an attempt for more intelligent read/write caching: http://www.microsoft.com/whdc/driver/priorityio.mspx --- .je ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re[2]: [zfs-discuss] Zones root-fs in ZFS ? (fwd)
Hello Lori, Friday, July 28, 2006, 6:50:55 PM, you wrote: >>> Right now I have such configurations and have been using smpatch >>> without any problems so far. >> >> I thought I read somewhere (zones guide?) that putting the zone root fs >> on zfs was unsupported. LA> You've missed the earlier part of this thread. Yes, it's LA> unsupported, but the question was asked "Does it work anyway?" LA> and the short answer is "yes, it mostly does, but there LA> are enough complications and known problems with the LA> patching and upgrade of this configuration that it LA> is currently unsupported." Not that much problems actually. If you are not going to use Live Upgrade and you put only zone's root-fs on a ZFS and do not create another ZFS filesystems for zone's /var, /opt, /usr, etc. then you shouldn't be affected. In such a config smpatch/updatemanager just work. -- Best regards, Robertmailto:[EMAIL PROTECTED] http://milek.blogspot.com ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Re: [zfs-discuss] ZFS needs a viable backup mechanism
On Thu, Jul 27, 2006 at 11:46:30AM -0700, Richard Elling wrote: > >>I'm don't have visibility of the Explorer development sites at the > >>moment, but I believe that the last publicly available Explorer I > >>looked at (v5.4) still didn't gather any ZFS related info, which would > >>scare me mightily for a FS released in a production-grade Solaris 10 > >>release ... how do we expect our support personnel to engage?? > > Timing is everything :-) > http://docs.sun.com/app/docs/doc/819-6612 Timing is indeed everything - and it's the reason we didn't have ZFS support until now. The Explorer team has been following ZFS for over 2 years (CR 5074463 to add support was created 17 July 2004!), but made the decision long ago that we would wait until we were sure exactly what we should be doing in Explorer before we actually added the relevant support. As it turned out, this was a good thing - I don't think there's a single command that was originally listed in the CR that still exists due to the changes that occured as ZFS matured to the point of release. Explorer 5.5 is the first release since Solaris 10 6/06, and thus the one we ended up adding ZFS support into. It also has a few other new features, including the ability to automatically transfer an Explorer to Sun over HTTPS. Scott. ___ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
