toasters February 2007

toasters@lists.teaparty.net

50 participants
33 discussions

Re: data management through compression for Netapp
by Wilbur Castro 27 Feb '07

27 Feb '07

I took a look at storewiz. It sure looks promising. How transparent is it though? Do the nfs and windows clients know that the storewiz appliance exists (proxy) and do clients need to be reconfigured? or do they do some packet switching type of things? How does this work with your snapshots? Frequently changing files may take up more space since they are snapshotted. Does this work well for random access types of applications? I will need to test this thing like hell; it's either a promotion for cutting storage costs, or I lose my job if we corrupt data ;-) Thanks for the pointers guys. On 2/26/07, Skottie Miller <Scott.Miller(a)dreamworks.com> wrote: > > > > Wilbur Castro wrote: > > Hi toasters, > > > > We have a couple of 100 TBs of heterogenous storage (some from netapp), > > and see that our storage grows close to 60% over year. We were looking > > at alternatives for managing this data growth. Compression was one of > > the techniques we were considering for our nearline and (possibly) > > primary storage. Our applications cannot change to do their own > > compression, so it boils down to doing this in the storage layer or > > through an external device. Also, we'd like to not have any performance > > impact and compression to happen transparently. Deduplication technology > > from storage vendors would help, but it is not a hetrogenous solution. > > > > I am not aware of any compression technology from netapp. Are you folks > > aware of any solutions? Would love to hear your experience with those or > > other alternative ways you deal with the storage growth problem while > > managing costs. > > Look at StoreWiz. > http://www.storewiz.com > > in-line, transparent boxes that go between your > filers and their switch ports. > > I'm evaluating them now, and so far, so good. > > They have a software app that emulates their hardware, > so you can get a sense of the compression that will > result on your data set. > > also look at Acopia (http://www.acopia.com), they make > an "NFS switch" that lets you do transparent tiering > and data migration w/o impact to the applications. > consider a deployment where the "hot" few TB of data > are on the fastest/costliest filers, and the rest > is on lower-cost second-tier storage. put the acopia > logicially between the filers and the apps, and you > get alot of options, and the apps don't even know. > > we have one in-place for some long-term exposure testing, > and are considering more for some future roll-outs. > > -skottie > > > > > Thx, > > Wilbur > > -- > > Scott Miller > skottie(a)anim.DreamWorks.com >

2 1

Software Ownership Query
by Philip Boyle 22 Feb '07

22 Feb '07

Hi guys, I've just received an FAS6030 cluster. The system shipped with ONTAP 7.2. I've upgraded the system to 7.2.1.1 bfiler09> version -b 1:/x86_64/kernel/primary.krn: OS 7.2.1.1 1:/backup/x86_64/kernel/primary.krn: OS 7.2 1:/x86_64/diag/diag.krn: 4.7 1:/x86_64/firmware/excelsio/firmware.img: Firmware 1.2.0 1:/x86_64/firmware/DrWho/firmware.img: Firmware 2.0.0 1:/boot/loader: Loader 1.2.1 Do you guys have much experience of disk ownership. On our FAS940 clusters, disk ownership was hardware based. FAS6030> disk show -v DISK OWNER POOL SERIAL NUMBER ------------ ------------- ----- ------------- 0e.40 FAS6030 (118041055) Pool0 ZAKTSGAH Tue Feb 20 11:13:43 GMT [localhost: diskown.isEnabled:info]: replayed event: software ownership has been enabled for this system Despite reading the below Netapp doc, I'm no closer to making a decision as to whether to retain the disk ownership option or remove it from my system. http://now.netapp.com/NOW/knowledge/docs/ontap/rel7211/ <http://now.netapp.com/NOW/knowledge/docs/ontap/rel7211/html/ontap/mgmtsag/3…> html/ontap/mgmtsag/3disk6.htm#1426071 <http://now.netapp.com/NOW/knowledge/docs/ontap/rel7211/html/ontap/mgmtsag/3…> Any feedback on the topic would be much appreciated. Thanks alot, Philip

2 2

Quota CPU hog with Ontap 7.0
by JONES P.S. 21 Feb '07

21 Feb '07

We've just upgraded our main filers to 7.0.5 after running 588 days with no interruption on 6.5.5. However we've run into serious problems when enabling quotas. The system runs at 100% for ages and we eventually have to disable quotas altogether. On the old system it took about 1 hour and 20% CPU to initialise quotas so there's an enormous difference. The system is unuseable if we enable quotas on 7.0.5. Netapps haven't been much help; the articles they refer me too mention this problem but in all cases the bug notes claim its been fixed in 7.0.5. Anyone been here before me? We have ~20,000 accounts but most are captured with default quotas although we have around 1000 explicit entries defined in the quotas file. Thanks ... Paul Paul Jones, B.Sc, M.Sc, Ph.D. - Head of Systems Services, Information Technology Service, Durham University, South Road, Durham DH1 3LE Email: p.s.jones(a)durham.ac.uk Phone/Fax: 0191 334 2749/2701

2 1

Re: NFSv3 + krb5 home directory problem
by Jim Davis 20 Feb '07

20 Feb '07

Brian Parent wrote: > I haven't seen any responses yet. Have you opened a ticket with NetApp? No; since the mount itself seems to succeed I think the Netapp side is doing the right thing. Why the credentials aren't being passed and why rpc.gssd isn't seeing anything is still a mystery. Posting to the Kerberos list produced no responses; posting to the Linux NFS list got a reply from one of the developers, but no fixes yet. > Please post a summary back to the list if you ever do make any progress. Big "if"... but I'll be glad to do so.

8 7

RE: Large Oracle database question
by Miller, Michael CTR USTRANSCOM J2 20 Feb '07

20 Feb '07

I agree with your thoughts on performance but not the use of large raid groups. It would eat into the data spindle count but I think 10 disk RGs would increase the warm fuzzy feeling. I have pushed single parity RGs well past the safe point without problems but the guy is antsy about failures. I have never lost any data in the years I have been taking care of the filers. The plan right now is to use what comes with Solaris for filesystem management. VxVM was thought of for a while. Mike -----Original Message----- From: slinkymax0r [mailto:slinkywizard@integra.net] Sent: Sunday, February 18, 2007 4:06 PM To: tmac Cc: Miller, Michael CTR USTRANSCOM J2; toasters(a)mathworks.com Subject: Re: Large Oracle database question hmmm... with an adequately sized head, his work load shouldn't be a problem at all. I would actually anticipate better performance/ recoverability with large aggregates of fast spindles, and RAID-DP means very large raid groups :) Would you be using NFS or block? If block, what file system would you be using over the RAID array? Regards, Max > Has anyone explained or showed this person a write-up on WAFL and > NetApp's implementation of RAID-4 and now RAID-DP? > ...how NetApp's version does not have a "hot" parity disk as all disks > are excersized about the same? > ...how RAID-DP is superior to RAID 1+0 and in an edge case is even > better (if you loose the same disk in both plexs of a RAID 1, you have > total data loss...you can loose any two disks with RAID-DP and loose > nothing) > ...how RAID-DP has very little overhead compared to NetApp's RAID-4? > ...how NetApp's RAID-4 is faster than most hardware RAID arrays? > > > I could go on, and on, an on....(ask others that know me...plus 10 > years at netapp helped) > > -- > --tmac > > RedHat Certified Engineer > > > > On 2/16/07, Miller, Michael CTR USTRANSCOM J2 > <michael.miller.ctr(a)ustranscom.mil> wrote: >> >> >> >> (This might show up twice) >> We are getting ready to stand up a 30+TB GIS app and I was wondering if >> anyone on the list has done something like this on a filer. >> >> The guy who will be running this doesn't think a filer would give him >> the >> performance required (not specified) and is also concerned that the >> parity >> drives will get beat to hard. >> >> He has been told by others that the only way to make this work as needed >> is >> to use RAID 1+0. This started out being required for all of the data but >> has >> been relaxed to just the indexes and current point of interest data. >> >> Any thoughts? >> >> >> Mike Miller >> General Dynamics Information Technology >> Michael.Miller.ctr(a)ustranscom.mil >> Michael.Miller(a)gdit.com >> Phone: 618-229-1185 >> >

1 0

F760 memory spec
by Ronan Mullally 19 Feb '07

19 Feb '07

For sheer geek purposes, I've got an old F760, long out of warranty, in my garage. One of the DIMMs has developed a fault and I'd like to replace it. Does anybody know what the memory spec is? The DIMM is branded Samsung KM48S16030BT-GL. From what I can gather it's 168 pin 256MB SDRAM, but I don't know what speed (PC-100? PC-133?), and given that it's got 9 chips on it, it's ECC. Is the filer fussy about the memory it will accept? Do I have to replace this with identical memory, or will a pair of similar DIMMS do? -Ronan

1 0

DST files
by admin＠berkcom.net 17 Feb '07

17 Feb '07

5 5

Large Oracle database question
by Miller, Michael CTR USTRANSCOM J2 16 Feb '07

16 Feb '07

(This might show up twice) We are getting ready to stand up a 30+TB GIS app and I was wondering if anyone on the list has done something like this on a filer. The guy who will be running this doesn't think a filer would give him the performance required (not specified) and is also concerned that the parity drives will get beat to hard. He has been told by others that the only way to make this work as needed is to use RAID 1+0. This started out being required for all of the data but has been relaxed to just the indexes and current point of interest data. Any thoughts? Mike Miller General Dynamics Information Technology Michael.Miller.ctr(a)ustranscom.mil Michael.Miller(a)gdit.com Phone: 618-229-1185

4 3

BayLISA meeting tonight - Feb 15 7-10pm NetApp HQ Sunnyvale
by Jennifer Davis 16 Feb '07

16 Feb '07

I know this is late notice, but I forgot that there might be people interested on this forum in this talk. If you are interested, make sure to RSVP: http://www.eventbrite.com/event/46941403 We will be recording the meeting, so if you can't make it in person you can always watch the video recording after the fact. The February General meeting will be held at NetApp HQ. Location: Network Appliance HQ Building 1, Founders' Theater 495 E Java Dr Sunnyvale, CA 94089 Time: 7:00-10:00pm Architecting a Resilient Storage System Steve Strange, NetApp Engineering The job of a storage engineer is to find ways to turn a collection of unreliable disks into a resilient system. In this presentation, NetApp engineer Steve Strange shares his perspective on some of the core technologies that help make NetApp systems reliable and easy to manage. In particular, Steve will discuss the NetApp RAID-DP implementation.which carries no significant performance penalty.versus traditional implementations of RAID-6, as well as other alternative approaches to storage problems. Steve Strange has been a member of the NetApp engineering team for nine years and has worked on numerous resiliency projects including the launch of NetApp's first clustered failover product, the rearchitecture of the RAID subsystem, and the development of SyncMirror. Case Study: Decreasing Data Center Power Utilization Devinder Singh, NetApp IT Architect In a recent upgrade of its corporate datacenter, NetApp IT reduced the physical footprint of its storage by 40% and cut power consumption by up to 75% while doubling performance. All told, the project touched 165 different applications (with as many as 8 environments each), 343 servers, and just under 5000 mounts. In this presentation, IT Architect Devinder Singh describes the reasons for the upgrade, the migration methodology, and the results.

2 1

High disk reads but low network out
by Paul Letta 15 Feb '07

15 Feb '07

I have been having some performance issues on my F880 the past couple of days. Specifically, an NDMP backup that normally takes 2 hours to complete was stuck in the mapping phase for 12 hours before I killed it. What I am seeing is a high disk read, but without a corresponding network out. Here is a sysstat: > CPU NFS CIFS HTTP Net kB/s Disk kB/s Tape kB/s Cache > in out read write read write age > 62% 1456 6 0 794 16954 46326 0 0 0 4s > 56% 1418 0 0 597 13689 36506 0 0 0 5s > 64% 1737 3 0 867 17637 43572 0 0 0 5s > 53% 1135 18 0 572 14823 42905 1155 0 0 4s > 73% 2067 20 0 1282 37835 69880 6347 0 0 4s > 69% 2031 24 0 1267 38868 67632 0 0 0 4s > 74% 1921 3 0 1027 26644 50881 7778 0 0 5s > 59% 1459 0 0 774 18843 41313 0 0 0 6s > 60% 1212 2 0 665 16391 38766 0 0 0 7s > 54% 1028 19 0 509 11873 33960 0 0 0 7s > 61% 844 302 0 571 13765 39032 0 0 0 7s > 72% 1164 598 0 1870 16750 39968 0 0 0 5s > 71% 948 640 0 8961 9788 29552 0 0 0 5s > 79% 1297 603 0 24157 5831 26034 0 0 0 4s > 87% 2130 465 0 23581 9794 31269 0 0 0 2s I don't know why the disks are reading so much without it going out the network. This is an F880 running OnTap 7.1. It is somewhat full -- about 92% (1.2TB out of 1.3TB). There isn't any stuck ndmp sessions: > ndmpd status ndmpd ON. No ndmpd sessions active. I haven't really looked at what clients may be doing what due to the low network out numbers- but I am going there next. Does anyone have any suggestions on what to look for ? Thanks, Paul

4 4

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

toasters February 2007