toasters October 2006

toasters@lists.teaparty.net

30 participants
28 discussions

RE: Estimating Aggregate IOPS
by Glenn Walker 05 Oct '06

05 Oct '06

That's not just a NetApp thing... that's performance 101! There will ALWAYS be a bottleneck somewhere - if you want to 'fix' a performance problem, find the first bottleneck, and fix it. Then the next, and so on. Keep going until the performance is adequate... I've seen very few cases where the loop itself gets saturated before the head gives up, or the memory runs out... but it can happen. I expect that it will happen even more quickly with the new 6000 series given the sheer amount of memory\nvram\cpu. Then again, 500TB of disks is a LOT of loop too :) Glenn -----Original Message----- From: Brosseau, Paul [mailto:Paul.Brosseau@netapp.com] Sent: Wednesday, October 04, 2006 8:17 PM To: Glenn Walker; Suresh Rajagopalan Cc: toasters(a)mathworks.com Subject: RE: Estimating Aggregate IOPS Good point Glenn on the number of disks. Once the number of IOPS on a loops generates enough throughput to saturate the loop there's not much point in adding more disks. Spreading the load across multiple loops helps. How does the adage go " go wide then go shallow then go deep ". Eventually you saturate something somewhere depending on the filer model . Paulb -----Original Message----- From: Glenn Walker [mailto:ggwalker@mindspring.com] Sent: Wednesday, October 04, 2006 6:57 PM To: Suresh Rajagopalan Cc: toasters(a)mathworks.com Subject: RE: Estimating Aggregate IOPS There is a ceiling in performance gain after a certain number of disks - it's well below 112, but I'm pretty sure it's above 56 (though not much above it). At that point, the 'curve' starts to go back down again. However, RAID-DP and RAID-4 (it's poor, homely cousin) are very fast and the filer has always handled the XORs very well for the parity calculations. A better question\way to position this might be what type of workload\performance you expect\require. While the disks are capable of about 120 IOPS for 10k and 180 IOPS for 15K disks, the latencies begin to jump at the end of the scale - if you're workload requires lower latencies, it doesn't much matter what the disk is capable of (Exchange is a very good example of this - 80 IOPS for 10k drives is about the max you'd want to go). Glenn -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Suresh Rajagopalan Sent: Wednesday, October 04, 2006 3:56 PM Cc: toasters(a)mathworks.com Subject: RE: Estimating Aggregate IOPS Let's assume 4k IOPS. Disks are currently 10k rpm. Are you saying that NTAP's implementation of RAID4 (or RAID-DP) gives a linear performance increase with spindles, (almost like RAID-0) and no penalty for the RAID4 or RAID-DP? That is, aggregate IOPS is pretty much (N * disk-iops) and no penalty for RAID4 or RAID-DP? Thanks Suresh -----Original Message----- From: Blake Golliher [mailto:thelastman@gmail.com] Sent: Wednesday, October 04, 2006 12:41 PM To: Suresh Rajagopalan Cc: toasters(a)mathworks.com Subject: Re: Estimating Aggregate IOPS How fast are the spindles, and what size is your iop? Generally, you can assume 110 per 10k RPM disk, and 180 for 15k RPM disk. This is also assuming a 10ms latency for each iop. And subtract 2 disks per raid group for raid dp (if doing writes, don't if you are doing pure reads). So with around 50 disks (after you subtract the raid dp overhead) you can expect around 5500 iops from that set of disks (assuming 10k rpm disks). For the 112 aggregate, you an expect 98 spindles and 10780 iops. I'm assuming 4k iops, and 10ms latency. Hope that helps, -Blake On 10/4/06, Suresh Rajagopalan <SRajagopalan(a)williamoneil.com> wrote: > Given a disk IOPS of 100, I'd like to estimate total aggregate IOPS for > the following cases: > > 1) 56 disks, 1 aggregate, RAID-DP size 16 > 2) 112 disks, 1 aggregate, RAID-DP size 16 > > I'm only interested in the total raw disk IOPS available in each case, > not including considering the filer head. For example, we know that > RAID-0 with 56 disks @100 would yield 5600 IOPS. > > I don't know how to do this calculation with Data ONTAP's implementation > of RAID4 or RAID-DP. > > Any assistance would help. > > Thanks > Suresh > > > > >

1 0

RE: Estimating Aggregate IOPS
by Brosseau, Paul 05 Oct '06

05 Oct '06

Good point Glenn on the number of disks. Once the number of IOPS on a loops generates enough throughput to saturate the loop there's not much point in adding more disks. Spreading the load across multiple loops helps. How does the adage go " go wide then go shallow then go deep ". Eventually you saturate something somewhere depending on the filer model . Paulb -----Original Message----- From: Glenn Walker [mailto:ggwalker@mindspring.com] Sent: Wednesday, October 04, 2006 6:57 PM To: Suresh Rajagopalan Cc: toasters(a)mathworks.com Subject: RE: Estimating Aggregate IOPS There is a ceiling in performance gain after a certain number of disks - it's well below 112, but I'm pretty sure it's above 56 (though not much above it). At that point, the 'curve' starts to go back down again. However, RAID-DP and RAID-4 (it's poor, homely cousin) are very fast and the filer has always handled the XORs very well for the parity calculations. A better question\way to position this might be what type of workload\performance you expect\require. While the disks are capable of about 120 IOPS for 10k and 180 IOPS for 15K disks, the latencies begin to jump at the end of the scale - if you're workload requires lower latencies, it doesn't much matter what the disk is capable of (Exchange is a very good example of this - 80 IOPS for 10k drives is about the max you'd want to go). Glenn -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Suresh Rajagopalan Sent: Wednesday, October 04, 2006 3:56 PM Cc: toasters(a)mathworks.com Subject: RE: Estimating Aggregate IOPS Let's assume 4k IOPS. Disks are currently 10k rpm. Are you saying that NTAP's implementation of RAID4 (or RAID-DP) gives a linear performance increase with spindles, (almost like RAID-0) and no penalty for the RAID4 or RAID-DP? That is, aggregate IOPS is pretty much (N * disk-iops) and no penalty for RAID4 or RAID-DP? Thanks Suresh -----Original Message----- From: Blake Golliher [mailto:thelastman@gmail.com] Sent: Wednesday, October 04, 2006 12:41 PM To: Suresh Rajagopalan Cc: toasters(a)mathworks.com Subject: Re: Estimating Aggregate IOPS How fast are the spindles, and what size is your iop? Generally, you can assume 110 per 10k RPM disk, and 180 for 15k RPM disk. This is also assuming a 10ms latency for each iop. And subtract 2 disks per raid group for raid dp (if doing writes, don't if you are doing pure reads). So with around 50 disks (after you subtract the raid dp overhead) you can expect around 5500 iops from that set of disks (assuming 10k rpm disks). For the 112 aggregate, you an expect 98 spindles and 10780 iops. I'm assuming 4k iops, and 10ms latency. Hope that helps, -Blake On 10/4/06, Suresh Rajagopalan <SRajagopalan(a)williamoneil.com> wrote: > Given a disk IOPS of 100, I'd like to estimate total aggregate IOPS for > the following cases: > > 1) 56 disks, 1 aggregate, RAID-DP size 16 > 2) 112 disks, 1 aggregate, RAID-DP size 16 > > I'm only interested in the total raw disk IOPS available in each case, > not including considering the filer head. For example, we know that > RAID-0 with 56 disks @100 would yield 5600 IOPS. > > I don't know how to do this calculation with Data ONTAP's implementation > of RAID4 or RAID-DP. > > Any assistance would help. > > Thanks > Suresh > > > > >

1 0

Re: Estimating Aggregate IOPS
by Blake Golliher 04 Oct '06

04 Oct '06

Glenn is very much correct here. The application will dictate what your upper limit will be. A highly sequential Oracle workload will different greatly from a 5000 user exchange cluster. That curve in raid groups is around 20 disks in my experience. Eventually you'll run out of head room for the number of disks you've attached. -Blake On 10/4/06, Glenn Walker <ggwalker(a)mindspring.com> wrote: > There is a ceiling in performance gain after a certain number of disks > it's well below 112, but I'm pretty sure it's above 56 (though not much > above it). At that point, the 'curve' starts to go back down again. > > However, RAID-DP and RAID-4 (it's poor, homely cousin) are very fast and > the filer has always handled the XORs very well for the parity > calculations. > > A better question\way to position this might be what type of > workload\performance you expect\require. While the disks are capable of > about 120 IOPS for 10k and 180 IOPS for 15K disks, the latencies begin > to jump at the end of the scale - if you're workload requires lower > latencies, it doesn't much matter what the disk is capable of (Exchange > is a very good example of this - 80 IOPS for 10k drives is about the max > you'd want to go). > > Glenn > > -----Original Message----- > From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] > On Behalf Of Suresh Rajagopalan > Sent: Wednesday, October 04, 2006 3:56 PM > Cc: toasters(a)mathworks.com > Subject: RE: Estimating Aggregate IOPS > > Let's assume 4k IOPS. Disks are currently 10k rpm. > > Are you saying that NTAP's implementation of RAID4 (or RAID-DP) gives a > linear performance increase with spindles, (almost like RAID-0) and no > penalty for the RAID4 or RAID-DP? > > That is, aggregate IOPS is pretty much (N * disk-iops) and no penalty > for RAID4 or RAID-DP? > > Thanks > Suresh > > > > -----Original Message----- > From: Blake Golliher [mailto:thelastman@gmail.com] > Sent: Wednesday, October 04, 2006 12:41 PM > To: Suresh Rajagopalan > Cc: toasters(a)mathworks.com > Subject: Re: Estimating Aggregate IOPS > > How fast are the spindles, and what size is your iop? Generally, you > can assume 110 per 10k RPM disk, and 180 for 15k RPM disk. This is > also assuming a 10ms latency for each iop. And subtract 2 disks per > raid group for raid dp (if doing writes, don't if you are doing pure > reads). So with around 50 disks (after you subtract the raid dp > overhead) you can expect around 5500 iops from that set of disks > (assuming 10k rpm disks). For the 112 aggregate, you an expect 98 > spindles > and 10780 iops. I'm assuming 4k iops, and 10ms latency. > > Hope that helps, > -Blake > > > On 10/4/06, Suresh Rajagopalan <SRajagopalan(a)williamoneil.com> wrote: > > Given a disk IOPS of 100, I'd like to estimate total aggregate IOPS > for > > the following cases: > > > > 1) 56 disks, 1 aggregate, RAID-DP size 16 > > 2) 112 disks, 1 aggregate, RAID-DP size 16 > > > > I'm only interested in the total raw disk IOPS available in each case, > > not including considering the filer head. For example, we know > that > > RAID-0 with 56 disks @100 would yield 5600 IOPS. > > > > I don't know how to do this calculation with Data ONTAP's > implementation > > of RAID4 or RAID-DP. > > > > Any assistance would help. > > > > Thanks > > Suresh > > > > > > > > > > > > >

1 0

RE: Maximum throughput of a FAS270c
by Glenn Walker 04 Oct '06

04 Oct '06

Yup... Pretty capable machine for 500 users or so. Depends on workload, as Blake said, it's OPS\CPU limited. 940 is a beast by comparison - probably about 2-2.5 times as fast. Glenn -----Original Message----- From: Blake Golliher [mailto:thelastman@gmail.com] Sent: Wednesday, October 04, 2006 3:44 PM To: John Stoffel Cc: Grey Friday; Glenn Walker; Brian Pascal; toasters(a)mathworks.com Subject: Re: Maximum throughput of a FAS270c I'd say an F840 (I think that's what Glenn was suggesting). The 270 is an op limited machine. Dual core 650Mhz processor, and a gig of ram (subtract 256 for NVMEM) and you get a pretty limited box in terms of performance. A 940 has a 1.8 ghz proc 4 or so gigs of memory, and a dedicated nvram card. Much better performance box. But a 270 wasn't marketed as a very fast box, but a very small, cheap box for small & remote offices. -Blake On 10/4/06, John Stoffel <john.stoffel(a)taec.toshiba.com> wrote: > >>>>> "Grey" == Grey Friday <grey.friday(a)gmail.com> writes: > > Grey> http://www.spec.org/sfs97r1/results/res2004q2/sfs97r1-20040525-00186.htm l > Grey> On 10/3/06, Glenn Walker <ggwalker(a)mindspring.com> wrote: > Grey> Brian, > >> > >> Those numbers are deemed NetApp Internal Only - though I believe that > >> Sales can share some info with you. > >> > >> However this might help: a FAS270c is equivalent to a NetApp F840 > >> system with respect to throughputhorsepower. Maybe just a tad faster > >> even. > > Glenn, > > Just to be sure, you're talking about an F840 is roughtly equivilent > to an FAS270, right? Not an F940? > > John > PM disk. This is also assuming a 10ms latency for each iop. And subtract 2 disks per raid group for raid dp (if doing writes, don't if you are doing pure reads). So with around 50 disks (after you subtract the raid dp overhead) you can expect around 5500 iops from that set of disks (assuming 10k rpm disks). For the 112 aggregate, you an expect 98 spindles and 10780 iops. I'm assuming 4k iops, and 10ms latency. Hope that helps, -Blake On 10/4/06, Suresh Rajagopalan <SRajagopalan(a)williamoneil.com> wrote: > Given a disk IOPS of 100, I'd like to estimate total aggregate IOPS for > the following cases: > > 1) 56 disks, 1 aggregate, RAID-DP size 16 > 2) 112 disks, 1 aggregate, RAID-DP size 16 > > I'm only interested in the total raw disk IOPS available in each case, > not including considering the filer head. For example, we know that > RAID-0 with 56 disks @100 would yield 5600 IOPS. > > I don't know how to do this calculation with Data ONTAP's implementation > of RAID4 or RAID-DP. > > Any assistance would help. > > Thanks > Suresh > > > > >

1 0

RE: Estimating Aggregate IOPS
by Suresh Rajagopalan 04 Oct '06

04 Oct '06

Let's assume 4k IOPS. Disks are currently 10k rpm. Are you saying that NTAP's implementation of RAID4 (or RAID-DP) gives a linear performance increase with spindles, (almost like RAID-0) and no penalty for the RAID4 or RAID-DP? That is, aggregate IOPS is pretty much (N * disk-iops) and no penalty for RAID4 or RAID-DP? Thanks Suresh -----Original Message----- From: Blake Golliher [mailto:thelastman@gmail.com] Sent: Wednesday, October 04, 2006 12:41 PM To: Suresh Rajagopalan Cc: toasters(a)mathworks.com Subject: Re: Estimating Aggregate IOPS How fast are the spindles, and what size is your iop? Generally, you can assume 110 per 10k RPM disk, and 180 for 15k RPM disk. This is also assuming a 10ms latency for each iop. And subtract 2 disks per raid group for raid dp (if doing writes, don't if you are doing pure reads). So with around 50 disks (after you subtract the raid dp overhead) you can expect around 5500 iops from that set of disks (assuming 10k rpm disks). For the 112 aggregate, you an expect 98 spindles and 10780 iops. I'm assuming 4k iops, and 10ms latency. Hope that helps, -Blake On 10/4/06, Suresh Rajagopalan <SRajagopalan(a)williamoneil.com> wrote: > Given a disk IOPS of 100, I'd like to estimate total aggregate IOPS for > the following cases: > > 1) 56 disks, 1 aggregate, RAID-DP size 16 > 2) 112 disks, 1 aggregate, RAID-DP size 16 > > I'm only interested in the total raw disk IOPS available in each case, > not including considering the filer head. For example, we know that > RAID-0 with 56 disks @100 would yield 5600 IOPS. > > I don't know how to do this calculation with Data ONTAP's implementation > of RAID4 or RAID-DP. > > Any assistance would help. > > Thanks > Suresh > > > > >

1 0

RE: Maximum throughput of a FAS270c
by Glenn Walker 04 Oct '06

04 Oct '06

Brian, Those numbers are deemed NetApp Internal Only - though I believe that Sales can share some info with you. However this might help: a FAS270c is equivalent to a NetApp F840 system with respect to throughput\horsepower. Maybe just a tad faster even. Glenn -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Brian Pascal Sent: Tuesday, October 03, 2006 1:37 AM To: toasters(a)mathworks.com Subject: Maximum throughput of a FAS270c Does any one has an idea of what is the maximum throughput of a FAS270c system? (Sustained system bandwidth & IOPS) Thanks, Brian. This Mail Has Been Scanned For Virus By Scanmail For Lotus Notes Enterprise/Kbsl

5 5

Estimating Aggregate IOPS
by Suresh Rajagopalan 04 Oct '06

04 Oct '06

Given a disk IOPS of 100, I'd like to estimate total aggregate IOPS for the following cases: 1) 56 disks, 1 aggregate, RAID-DP size 16 2) 112 disks, 1 aggregate, RAID-DP size 16 I'm only interested in the total raw disk IOPS available in each case, not including considering the filer head. For example, we know that RAID-0 with 56 disks @100 would yield 5600 IOPS. I don't know how to do this calculation with Data ONTAP's implementation of RAID4 or RAID-DP. Any assistance would help. Thanks Suresh

3 2

Maximum throughput of a FAS270c
by Brian Pascal 03 Oct '06

03 Oct '06

Does any one has an idea of what is the maximum throughput of a FAS270c system? (Sustained system bandwidth & IOPS) Thanks, Brian. This Mail Has Been Scanned For Virus By Scanmail For Lotus Notes Enterprise/Kbsl

1 0

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 October 2006