toasters

toasters@lists.teaparty.net

3 participants
13514 discussions

ok, so what do you use?
by dnb＠ccs.neu.edu 05 Aug '99

05 Aug '99

Howdy- The whole backup discussion leads me to a (potentially loaded) question: what software does NetApp (the company) use to back up its data? Peace, dNb

1 0

RE: Adding a FC shelf in a cluster w/o downtime
by Mohler, Jeff 05 Aug '99

05 Aug '99

My initial feelings are 'bad things will happen'. I dont have any technical data to back that up, other than its not documented to do it that way. -----Original Message----- From: Jay Soffian [mailto:jay@cimedia.com] Sent: Thursday, August 05, 1999 10:05 AM To: toasters(a)mathworks.com Subject: Adding a FC shelf in a cluster w/o downtime We are going to be adding an additional FC shelf to our clustered F740's. Obviously, the safe way to do this is to halt both 740's, attach the A & B loops, then bring both filers back up. But, what I was wonder was, would it be possible to: - Attach the B-loop first. - Have the B-Loop Filer take over all th shelves (including the new shelf). - Attach the A-Loop. - Give back the shelves to the A-Loop filer. I don't think we'll do this as we're only talking about about maybe 2 minutes of downtime vs 30 seconds, but I was wondering if this is a feasible thing to do. Thoughts? j.

1 0

RE: what's a good NFS write speed in MB/sec?
by Mohler, Jeff 05 Aug '99

05 Aug '99

THis is an interesting problem for NetApp. I purchased several NetApps for use on Oracle Servers for the Oracle DB. In evaluating all the info I too had problems relating nfs ops/sec to a disk subsystem throughput. --- Thats because you cant. NFS operations are many things, file creation, removal, buffering, data, lookups, info..etc. Writing a 1Gb file wont create nearly the operations that creating 1GB of 10k files. So, I've now got 2 F740 systems I'm looking at and so far like a lot. In using a single gigabit connection to a F740 I can easily get sequential writes of around 15mb/sec. At this point the Netapp is running CPU utilization of around +80% utilization. I've seen peaks of close to 18mb/s where the cpu utilization is pegged on 100%. --- This is about where you'll peg out Ive seen. Until the box has some serious CPU dedicated to RAID stripe calculations, and much MUCH more NVRAM..the platform is artificially limited. And if I understand correctly, your NVRAM caching becomes even less effective when you cluster. These are not BAD points about the filer at all, just places where improvements are taking place.

1 0

RE: Adding a FC shelf in a cluster w/o downtime
by Watanabe, Steven 05 Aug '99

05 Aug '99

> From: Jay Soffian [mailto:jay@cimedia.com] > We are going to be adding an additional FC shelf to our clustered > F740's. Obviously, the safe way to do this is to halt both 740's, > attach the A & B loops, then bring both filers back up. > > But, what I was wonder was, would it be possible to: > > - Attach the B-loop first. > - Have the B-Loop Filer take over all th shelves (including > the new shelf). > - Attach the A-Loop. > - Give back the shelves to the A-Loop filer. > > I don't think we'll do this as we're only talking about about maybe 2 > minutes of downtime vs 30 seconds, but I was wondering if this is a > feasible thing to do. Adding a disk shelf requires the FCAL cable to be unplugged while that shelf is being rewired. This requires the owning node to be HALTed at the time. It is possible to keep one member of the cluster alive during shelf rewiring. The approach is to HALT one node, failover to the other node, wire up the first node's A loop, and then bring that first node back to life. Then repeat the process, failing over and rewiring in the opposite direction.

1 0

RE: what's a good NFS write speed in MB/sec?
by Mohler, Jeff 05 Aug '99

05 Aug '99

Form an SGI Origin 2000 to Netapp F740, over 100Mbps Ethernet, I get 10MBytes/s consistent data transfer rate over nfs. However this is no good benchmark as the bottleneck here is ethernet speed and not filer. However if you can connect 3-4 NIC and get same performance an all four interfaces at the same time, this might suggest a very good performance. --- On a Sun U250, 2x300mhz, I can get about 80-90Mbits/sec if nothing is going on with the sun box. Multiple mounts, with multiple tests (into a switched, etherchannel environment to the F760) results in slower performance from the Sun. My next round of news servers in our cluster will be on non-PCI based hardware.

1 0

Re: what's a good NFS write speed in MB/sec?
by nathan＠dnase.hpw.pri.bms.com 05 Aug '99

05 Aug '99

My numbers from sgi origin2000 to sgi origin2000, with no router hops, came through at around 3Mbyte/sec over 100Mbit network and nfs. This was a for a transfer of a single 1GByte file. Interested in filer specs when communicating with origins, and also with gigabit ethernet hardware (presumably the origins can deal with the overhead?). nathan Matt Harrington <matt(a)ucsd.EDU> writes: > I'm evaluating a NAS filer similar to the ones sold by NetApp and Auspex. > This particular model is made by Nitech in Irvine, CA. These NFS vendors > speak about benchmarks in ops/sec. It's more intuitive for me to think of > benchmarks in terms of MB/sec throughputs for a large file write or read. > > My quick-and-dirty tests show that I can write at 1.8 MByte/sec over a > 100Mbit network using NFS v3. I can FTP at about 7 MByte/sec. > > My question is: do these numbers seem reasonable? If not, what aspects of > NFS should I tweak? > > ---matt > > > p.s. methods: > > to make a 100MB file: > dd if=/dev/zero of=/tmp/bigfile bs=1024 count=102400 > > time cp /tmp/bigfile /home/bigfile > > (where /home is the NFS server). > > -- N a t h a n O . S i e m e r s Bioinformatics Division of Applied Genomics Bristol-Myers Squibb Pharmaceutical Research Institute Hopewell Building 3B, P.O. Box 5400, Princeton, NJ 08543-5400 609 818-6568 siemersn(a)bms.com

2 1

Re: what's a good NFS write speed in MB/sec?
by Richard L. Rhodes 05 Aug '99

05 Aug '99

> Matt Harrington <matt(a)ucsd.EDU> writes: > > > I'm evaluating a NAS filer similar to the ones sold by NetApp and Auspex. > > This particular model is made by Nitech in Irvine, CA. These NFS vendors > > speak about benchmarks in ops/sec. It's more intuitive for me to think of > > benchmarks in terms of MB/sec throughputs for a large file write or read. > > > > My quick-and-dirty tests show that I can write at 1.8 MByte/sec over a > > 100Mbit network using NFS v3. I can FTP at about 7 MByte/sec. > > > > My question is: do these numbers seem reasonable? If not, what aspects of > > NFS should I tweak? > > > > ---matt THis is an interesting problem for NetApp. I purchased several NetApps for use on Oracle Servers for the Oracle DB. In evaluating all the info I too had problems relating nfs ops/sec to a disk subsystem throughput. NetApp wanted us to purchase a system on 30 day trial to see if it would do what I needed. I refused and told them that without an up front proof of concept there was no discussion of a trial purchase. So, our local tech rep brought a F6xx model (I forget) that we played with for half a day. With this box and 100BaseT connections to our RS/6k systems I got sequential writes of around 8-9mb/sec. This was much faster than one raid system I have that can only handle sequential writes at around 5mb/second, but not anywhere near the speed of a Hitachi Raid system I tested that could accept sequential writes at 25mb/sec (it had other problem that made it unusable). So, I've now got 2 F740 systems I'm looking at and so far like a lot. In using a single gigabit connection to a F740 I can easily get sequential writes of around 15mb/sec. At this point the Netapp is running CPU utilization of around +80% utilization. I've seen peaks of close to 18mb/s where the cpu utilization is pegged on 100%. What I've also been doing is running some standard benchmark tests against the netapp. While this provides very rough info, it's better than nfs ops/sec. Below are some results from a program called stkio ( http://www.stortek.com/StorageTek/software/freeware ): ----------------------------------------------------------- -- 1 process ---------------------------------------------- ----------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 1 IO type is seq writes file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 16.53 MB per second for 4KB (4232 IO's per second). Slowest IO 42429.75ms, Quickest IO 0.07ms, Mean 0.22ms 16.52 MB per second for 8KB (2114 IO's per second). Slowest IO 49188.87ms, Quickest IO 0.12ms, Mean 0.46ms 13.53 MB per second for 16KB (865 IO's per second). Slowest IO 39593.59ms, Quickest IO 0.24ms, Mean 1.13ms 17.48 MB per second for 32KB (559 IO's per second). Slowest IO 47646.34ms, Quickest IO 0.48ms, Mean 1.76ms ----------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 1 IO type is seq reads file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 15.97 MB per second for 4KB (4087 IO's per second). Slowest IO 2316.58ms, Quickest IO 0.03ms, Mean 0.23ms 18.54 MB per second for 8KB (2373 IO's per second). Slowest IO 370.41ms, Quickest IO 0.05ms, Mean 0.40ms 16.96 MB per second for 16KB (1085 IO's per second). Slowest IO 1754.96ms, Quickest IO 0.11ms, Mean 0.90ms 16.68 MB per second for 32KB (533 IO's per second). Slowest IO 416.97ms, Quickest IO 0.28ms, Mean 1.84ms ----------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 1 IO type is seq reads and writes file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 14.10 MB per second for 4KB (3609 IO's per second). Slowest IO 30118.43ms, Quickest IO 0.07ms, Mean 0.26ms 13.60 MB per second for 8KB (1740 IO's per second). Slowest IO 41456.73ms, Quickest IO 0.15ms, Mean 0.56ms 6.21 MB per second for 16KB (397 IO's per second). Slowest IO 2302.37ms, Quickest IO 0.42ms, Mean 2.49ms 6.99 MB per second for 32KB (223 IO's per second). Slowest IO 2177.81ms, Quickest IO 0.91ms, Mean 4.42ms ----------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 1 IO type is random writes file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 1.60 MB per second for 4KB (410 IO's per second). Slowest IO 38207.61ms, Quickest IO 0.03ms, Mean 2.42ms 4.34 MB per second for 8KB (554 IO's per second). Slowest IO 32236.26ms, Quickest IO 0.05ms, Mean 1.79ms 8.53 MB per second for 16KB (546 IO's per second). Slowest IO 39446.70ms, Quickest IO 0.11ms, Mean 1.82ms 11.47 MB per second for 32KB (367 IO's per second). Slowest IO 49895.58ms, Quickest IO 0.21ms, Mean 2.71ms ---------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 1 IO type is random reads file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 2.26 MB per second for 4KB (579 IO's per second). Slowest IO 635.95ms, Quickest IO 0.03ms, Mean 1.71ms 3.25 MB per second for 8KB (416 IO's per second). Slowest IO 553.14ms, Quickest IO 0.04ms, Mean 2.39ms 4.45 MB per second for 16KB (284 IO's per second). Slowest IO 545.42ms, Quickest IO 0.11ms, Mean 3.49ms 5.86 MB per second for 32KB (187 IO's per second). Slowest IO 533.81ms, Quickest IO 0.21ms, Mean 5.31ms ----------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 1 IO type is random reads and writes file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 1.55 MB per second for 4KB (397 IO's per second). Slowest IO 17135.13ms, Quickest IO 0.05ms, Mean 2.50ms 1.33 MB per second for 8KB (170 IO's per second). Slowest IO 17548.93ms, Quickest IO 0.10ms, Mean 5.84ms 1.47 MB per second for 16KB (93 IO's per second). Slowest IO 20197.14ms, Quickest IO 0.23ms, Mean 10.62ms 3.81 MB per second for 32KB (121 IO's per second). Slowest IO 24408.55ms, Quickest IO 0.47ms, Mean 8.18ms ----------------------------------------------------------- = 2 processes =============================================== ----------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 2 IO type is seq writes file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 14.07 MB per second for 4KB (3602 IO's per second). Slowest IO 3640.43ms, Quickest IO 0.04ms, Mean 0.54ms 14.08 MB per second for 8KB (1802 IO's per second). Slowest IO 6971.74ms, Quickest IO 0.06ms, Mean 1.09ms 14.32 MB per second for 16KB (916 IO's per second). Slowest IO 7378.41ms, Quickest IO 0.24ms, Mean 2.16ms 15.34 MB per second for 32KB (490 IO's per second). Slowest IO 14434.84ms, Quickest IO 0.45ms, Mean 4.04ms ----------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 2 IO type is seq reads file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 13.73 MB per second for 4KB (3519 IO's per second). Slowest IO 377.84ms, Quickest IO 0.03ms, Mean 0.55ms 17.47 MB per second for 8KB (2237 IO's per second). Slowest IO 447.69ms, Quickest IO 0.04ms, Mean 0.87ms 14.29 MB per second for 16KB (915 IO's per second). Slowest IO 451.40ms, Quickest IO 0.10ms, Mean 2.15ms 14.06 MB per second for 32KB (450 IO's per second). Slowest IO 418.53ms, Quickest IO 0.21ms, Mean 4.40ms ----------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 2 IO type is seq reads and writes file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 2.00 MB per second for 4KB (512 IO's per second). Slowest IO 1886.46ms, Quickest IO 0.07ms, Mean 3.88ms 2.93 MB per second for 8KB (375 IO's per second). Slowest IO 2520.66ms, Quickest IO 0.13ms, Mean 5.29ms 5.03 MB per second for 16KB (321 IO's per second). Slowest IO 353.10ms, Quickest IO 0.25ms, Mean 6.17ms 5.55 MB per second for 32KB (177 IO's per second). Slowest IO 1252.26ms, Quickest IO 0.59ms, Mean 11.21ms ----------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 2 IO type is random writes file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 1.65 MB per second for 4KB (422 IO's per second). Slowest IO 27679.15ms, Quickest IO 0.05ms, Mean 4.72ms 4.54 MB per second for 8KB (581 IO's per second). Slowest IO 32073.85ms, Quickest IO 0.09ms, Mean 3.42ms 8.55 MB per second for 16KB (547 IO's per second). Slowest IO 37631.32ms, Quickest IO 0.16ms, Mean 3.64ms 11.54 MB per second for 32KB (369 IO's per second). Slowest IO 48069.40ms, Quickest IO 0.33ms, Mean 5.40ms ---------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 2 IO type is random reads file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled 5.81 MB per second for 4KB (1488 IO's per second). Slowest IO 648.85ms, Quickest IO 0.03ms, Mean 1.33ms 6.98 MB per second for 8KB (894 IO's per second). Slowest IO 1760.59ms, Quickest IO 0.05ms, Mean 2.22ms 8.78 MB per second for 16KB (561 IO's per second). Slowest IO 610.85ms, Quickest IO 0.10ms, Mean 3.54ms 11.62 MB per second for 32KB (372 IO's per second). Slowest IO 609.94ms, Quickest IO 0.22ms, Mean 5.35ms ----------------------------------------------------------- STKIO Version 1.0 Copyright 1997 StorageTek Corporation Run for 1 loops to 1 files/volumes number of processes is 2 IO type is random reads and writes file size is 2000 MB Initial transfer length = 4096 bytes Ending Transfer length will be 32768 bytes I/O processes terminate after 2097152000 bytes are transferred Statistics capture enabled ---------------------------------------------------------------------- Richard L. Rhodes e: rhodesr(a)firstenergycorp.com Ohio Edison Co. p: 330-384-4904 f: 330-384-2514

1 0

Re: NetApp/NDMP
by sirbruce＠ix.netcom.com 05 Aug '99

05 Aug '99

On 08/04/99 15:27:17 you wrote: > NDMP is the strategic direction for backup at > Network Appliance. We see NDMP as a winner for Network > Attached Storage (NAS) vendors and ultimately we > think this could also be very prevalent in Storage > Area Network (SAN) environments too. When Netapp has a hybrid NAS/SAN solution, I would expect straight "local" backup to a tape device on the SAN to be an option, via traditional dump, without NDMP involved. Bruce

1 0

NetApp/NDMP
by Marchi, Michael 05 Aug '99

05 Aug '99

Bruce, NDMP is the strategic direction for backup at Network Appliance. We see NDMP as a winner for Network Attached Storage (NAS) vendors and ultimately we think this could also be very prevalent in Storage Area Network (SAN) environments too. There is definite momentum building for the use of NDMP. Back-up solutions for NetApp which use NDMP are now offered from four large vendors and we see third party support for additional solutions expanding. The currently offered solutions include Legato (BudTool), Veritas (NetBackup), CommVault (ABARS)& Workstation Solutions (Quick Restore). Note however that NetApp's solutions are in fact broader than NDMP and we are open to considering other alternatives. Other current solutions include - backing up over NFS mounts & CIFS shares, and - RPC-based solutions such as the about to be released Legato Networker Client for NetApp. You may not know it, but to achieve multiprotocol backup over CIFS with current (unchanged) NT backup products such as ARCserve & Backup Exec, NetApp engineering developed a patent pending solution that allows unix permissions to be wrapped into the ACLs captured by the NT backup process. As to the Legato Java-based client about to be released, Legato and NetApp worked jointly to develop this multi-protocol backup solution. The product is currently being tested at beta sites and should FCS early September. Bruce, NetApp is committed to NDMP and will continue to work with our partners to further define, enhance, and expand acceptance of NDMP as an industry standard. And we, as a company, will also continue to perfect our implementation. /***************************************** Mike Marchi - Director, Enterprise Marketing Network Appliance, Inc. (http://www.netapp.com) 495 East Java Drive Sunnyvale, CA 94089 Phone: (408) 822-6283 Fax: (408) 822-4415 Email: mjm(a)netapp.com *****************************************/

2 1

Re: 5.3D20 and bug #16084
by sirbruce＠ix.netcom.com 04 Aug '99

04 Aug '99

On 08/02/99 12:31:38 you wrote: > >Wow, I check my mail and in two consecutive e-mails I lose two clicks of >faith in NetApps. I think the overall consensus is that Netapp filers are still great, but there are a few particular areas where they are less than robust. NDMP is one of those (another is NIS). However, I wouldn't let that drive you away. Just choose a non-NDMP backup solution for the meantime. In a few years maybe NDMP will be perfect and everyone will support if fully. Or maybe some other new great standard will come along. Bruce

3 2

Jump to page:

2025

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