toasters

toasters@lists.teaparty.net

3 participants
13514 discussions

Re: SAN for SAS
by McDonald, Alex 11 Mar '11

11 Mar '11

Iirc (and I'd check since my info may be out of date) SAS 50/50 is only that over the long run; it does a lot of reads (up to entire analysis sets) followed by a lot of writes, with quite a bit of IO to temporary files too. It can be quite a handful to manage on shared storage. You need a good understanding of the app to be able to size it properly. Alex McDonald NetApp, Office of the CTO Send on a bb, excuse the typos ----- Original Message ----- From: Raj Patel <phigmov(a)gmail.com> To: Maxwell Reid <max.reid(a)saikonetworks.com> Cc: toasters(a)mathworks.com <toasters(a)mathworks.com> Sent: Fri Mar 11 20:16:12 2011 Subject: Re: SAN for SAS Hi Max, There are a couple of issues for these guys - * Long copy times for the datasets from the data-center to their workstations * Long processing times for i/o intensive SAS processes on their PC - they're just using SAS Workstation and batching up the work themselves (ie not distributed by design but by necessity) WAN circuit costs are pricey so ideally we'd centralise their workloads - powerful SAS server plus plenty of disk or even 40 servers with individual workstation licenses (either VDI, Citrix or real) and shared disk. Their theory is that their tasks aren't necessarily CPU bound but disk i/o bound. SSD's are fine for reads but they're skeptical about writes. SAS apparently supports RAM disks but they're pretty pricey for the size they'd need. A bit of googling indicates SAS has a distributed processing mechanism. I'll have to chat to them about licensing (suspect its not cheap). Anyone using their SAN's for storing or running weather, population or financial simulations (ie massive data-sets with a 50/50 mix of reads/writes) ? Cheers to all for the tips so far ! Raj. On 3/12/11, Maxwell Reid <max.reid(a)saikonetworks.com> wrote: > HI Raj, > > Would I be correct in assuming that the main problem you're trying to solve > is reducing the replication via WAN to the local machines? > > In this case, moving the processing nodes to 40 machines in the datacenter > with or without a shared storage arrangement would fix that issue; RPC or > protocol based replication between nodes would be required unless you're > going to switch to a clustered filesystem or network filesystem. In this > case a SAN or NAS setup would work fine. > > From the HPC side of things, sticking a bunch of PCIe SSDs in the nodes and > connecting them together via 10GbE or Infiniband would certainly speed > things up without the need for a shared disk pool, but that's probably > overkill for what you're trying to accomplish. > > > ~Max > > > > > On Fri, Mar 11, 2011 at 10:43 AM, Raj Patel <phigmov(a)gmail.com> wrote: > >> Hi, >> >> A bit of a generic SAN question (not necessarily NetApp specific). >> >> I've got a team of 40 people who use a statistical analysis package >> (SAS) to crunch massive time-series data sets. >> >> They claim their biggest gripe is disk contention - not necessarily >> one person using the same data but 40. So they process these data-sets >> locally on high-spec PC's with several disks (one for OS, one for >> scratch, one for reads and one for writes). >> >> I think they'd be much better off utilising shared storage (ie a SAN) >> in a datacenter so at least the workloads are spread across multiple >> spindles and they only need to copy or replicate the data within the >> datacenter rather than schlep it up and down the WAN which is what >> they currently do to get it to their distributed team PC's. >> >> Are there any useful guides or comparisons for best practise in >> designing HPC environments on shared infrastructure ? >> >> Other than knowing what SAS does I'm not sure on its HPC capabilities >> (ie distributed computing, clustering etc) so I'll need to follow that >> angle up too. >> >> Thanks in advance, >> Raj. >> >

1 0

Do I need to update diagnostics firmware?
by Ray Van Dolson 10 Mar '11

10 Mar '11

About to bump my N3600 A20 (FAS 2050C) from Data ONTAP 7.3.1 to 7.3.5.1. Our BIOS and BMC are current at 3.0 and 1.2 respectively, but our diagnostics firmware is at 5.3.6 (the latest is 5.4.6). Couple of questions: - Should I really update the Diagnostics firmware, or do many of you just leave it as-is if you're at the latest BIOS anyways? - Is it possible to update the Diagnostics firmware only (don't reflash the BIOS, etc) and somehow avoid needing to fail-over to my other controller and reboot? Thanks! Ray

3 2

Deduplication limit on 2040
by Milazzo Giacomo 10 Mar '11

10 Mar '11

Hi all, on FAS2040 with DoT 7 the size limit for a volume with A-SIS enabled is 3 TB. I read that with DoT 8.0.1 this limit has been extended to 16 TB (note, the limit with A-SIS enabled, not the vol size limit!) but nowhere I find that this can be done ONLY if containing aggregate is 64 bit. So, question is: to have a deduplicated volume on a 2040 bigger that 3 TB and up to 16 TB the containing aggregate has to be 64 bit? Thanks, PS) I think so because with 32 bit aggregate you cannot dedup a 16 TB vol (max size of the vol) because in this case no space is left on volume/aggr to dedup.

1 0

SAN data migration
by Milazzo Giacomo 06 Mar '11

06 Mar '11

Hi all, once upon a time there was "ReplicatorX"...and now? :) Anyway I've never used it and I don't know about its pros and cons. I should plan a SAN data migration from dozens of servers and hundreds of LUNs on HP EVA 5000, IBM DS 4800, EMX CX700 and 300 and so on... ...and all of this has to be performed online, with no disruption, better if agentless (os are Windows, Linux, AIX and more, cluster based on Windows, RHEL, Suse, Oracle...) Just one restart could be admitted for each systems... Any suggestion, tools ... Thanks in advance,

5 5

NFS Latency on VMware hosts
by Rob 16 Feb '11

16 Feb '11

Is there a way to measure the latency at the VMware host for an NFS mount? TIA -Rob

4 3

Moving FC ports to another SAN switch
by Steve Losen 15 Feb '11

15 Feb '11

Hello Toasters, We have a CF pair of fas6040s both connected to two SAN switches. We have migrated most of the clients off this SAN and now we have a need to connect the 6040s to a second (separate) SAN. Currently we have 2 SAN switches on the old SAN. Each 6040 head has two fc ports connected to each switch. So that is 4 fc ports per head or 8 ports total, which is complete overkill for this SAN (we plan to retire the few remaining clients.) I want to take one fc port from each head and move it to the other SAN, without any downtime. Can I do it like this (on each head): fcp config 0x down move the cable to the other switch fcp config 0x up We are using cfmode = single_image by the way. Is there anything else that we need to do on the netapp side? Steve Losen scl(a)virginia.edu phone: 434-924-0640 University of Virginia ITC Unix Support

1 0

cifs permissions doesnt match what i see in explorer
by Chaim Rieger 04 Feb '11

04 Feb '11

(last question for a while) in one of my cifs shares (and this applies to all of them) i have defined permissions as (example) domain/group1 Full Access domain/group2 Read Only however when i look at the security tab in windows explorer, it does not reflect as such. security style is the same across all the shares, (ntfs) and when i test the security by having various users access them they get access or denied based on what cifs shares state, not what is shown in explorer. i would like to have the explorer tab match, how can i accomplish this ?

4 3

Re: Growing an aggregate with different sized disks.
by joelk＠encs.concordia.ca 03 Feb '11

03 Feb '11

Maybe I should have been more explicit. Each raid group should be comprised of the same size/speed disks, and the number of disks in a raid group should be close to the recommended value for the type of drive and raid type. The only hiccup with building aggregates out of different size drives is when you want to replace the smaller older drives. You will have to migrate the whole aggregate (all the volumes) before you can destroy the aggregate and pull only the drives that you want to replace. joel Quoting Jeff Mohler <speedtoys.racing(a)gmail.com>: > Even so, different speeds can exist anywhere in the aggregate. > > The CP to disk is NOT complete until the last block is committed, no matter > where it is in the aggr...slow disks scattered across 4 RGs, or if all of > the slow ones are in one RG. Its a team effort. > > Raid stripes are only protection zones in disk failures, they are not > independent write zones where different disk speeds cause an explicit issue > when things are done correctly. > > > Now..if you create some small additional raid group at the end of an > existing aggregate, thats a different story, but not part of 'done > correctly' as stated above. > > > :) > > > > > > On Wed, Feb 2, 2011 at 3:01 PM, Joel Krajden <joelk(a)encs.concordia.ca>wrote: > >> You can specify the raid group size per raid group in an aggregate. As long >> as the speed of disks in the different raids groups are the same you should >> not see a performance issue for the aggregate. You should have at least one >> hot spare of each drive type and configure Ontap to use the same size drive >> when selecting a hot spare. Ontap should be at the latest revision if >> possible. >> >> >> Joel >> >> >> On 02/02/2011 05:24 PM, Ray Van Dolson wrote: >> >>> Have an aggregate comprised of 300GB disks and have two new shelves >>> with 450GB disks. >>> >>> I ran across this[1] which describes some concerns when dealing with >>> disks of mixed sizes. >>> >>> It sounds like I can't specify that 450GB disks and 300GB disks be >>> assigned to distinct raid groups (my raid size is 16), so if I add the >>> 450GB disks to my aggregate, I'll likely only be able to use 300GB of >>> them. >>> >>> Therefore I should create a new aggregate for the 450GB disks. >>> >>> Is my understanding of this correct? >>> >>> Also, the raid size seems to default to 16 -- I'm used to keeping >>> RAID-6 raid groups to 9 disks at most. Is 16 OK? >>> >>> Thanks, >>> Ray >>> >>> [1] >>> http://now.netapp.com/NOW/knowledge/docs/ontap/rel7261/html/ontap/smg/provi… >>> >> >> >> -- >> | Joel Krajden | Rm: EV-7105, Tel: 514 848-2424 3052 | >> | Core Technologies Mgr | | >> | Engineering & | Email: joelk(a)encs.concordia.ca | >> | Computer Science | | >> | Concordia University | In a circus, the clowns are supposed | >> | Montreal, Canada | to make you laugh, not cry. | >> > > > > -- > --- > Gustatus Similis Pullus >

1 0

maxdirsize question
by Chaim Rieger 02 Feb '11

02 Feb '11

is there a guideline for the maxdirsize setting based on the hardware ? or .... ?

2 1

Growing an aggregate with different sized disks.
by Ray Van Dolson 02 Feb '11

02 Feb '11

Have an aggregate comprised of 300GB disks and have two new shelves with 450GB disks. I ran across this[1] which describes some concerns when dealing with disks of mixed sizes. It sounds like I can't specify that 450GB disks and 300GB disks be assigned to distinct raid groups (my raid size is 16), so if I add the 450GB disks to my aggregate, I'll likely only be able to use 300GB of them. Therefore I should create a new aggregate for the 450GB disks. Is my understanding of this correct? Also, the raid size seems to default to 16 -- I'm used to keeping RAID-6 raid groups to 9 disks at most. Is 16 OK? Thanks, Ray [1] http://now.netapp.com/NOW/knowledge/docs/ontap/rel7261/html/ontap/smg/provi…

2 1

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