EMC and netapp iometer

I ran a few tests iometer on netapp vmdk and a cx380 vmdk and the netapp outperforms the vmdk cx380 by far. I don't know why that is.

I do not use powerpath on esxi servers so that there are 4 Active paths, only shows that IO active.

the netapp set to round robin and the two paths have active IO.

But why the netapp surpass the emc by so much?

the netapp is a 3210 on an aggregate with 14 disks dp.

the cx380 has 2 LUNS on 2 rgs

1 is a sata R5, 7 discs

2 is a CF R5 5 discs

both are using FCP

The number of the IOPS / s, you can do is directly proportional to the number of zones you throw.  The NetApp should bury the CX380 simply because the CX380 isn't enough time zones.  This is doubly true if the NetApp aggregate is the use of SAS drives.

Suppose that a disk SATA 7.2 K is capable of 100 IOPS / s and a SAS 15 K drive is capable of 180 IOPS / s.

If your 3210 has a aggregate of SATA drives, it's going up on 12 * 100 and 1,200 IOPS / s.  If they are SAS, the high-end is 2 160 IOPS / s.

Your RG1 CX380 will reach 600 and RG2 will reach 720.

It is not a fair controllers fight, since the controllers are much more capable IO, but you have forced the test by the number of batteries available.

Tags: VMware

Similar Questions

  • Storage - EMC or NetApp training? -What is the opinion which continue to increase identification of Collect information?

    Storage - EMC or NetApp training? -What is the opinion which continue to increase identification of Collect information?

    I get a lot of job opportunities for VCPs with storage capacities. I work with the directors of storage on all commitments but don't go or experience. Which storage certification I should continue to allow me to work as a PCV with extra storage responsibilities? My experience has been with EMC, NetApp and 3Parr on fiber, NFS, and iSCSI. What training and certification provider would be better or at least the first to continue?

    Thxs

    Both are huge players in the world of virtualization and thus increase your VCP - I would lean more towards the EMC cert just because of the VCE (VMware, CIsco and EMC) company

  • Migrrating a physical RDM to EMC for NetApp.

    I'm at the stage of planning for a replacement of SAN.  I need to find a painless and easy to migrate an RDM EMC for NetApp RDM.  He must remain in physical mode as it was created for a MS cluster in two boxes of ESXi.  The research I've read said that you must use tools such as the SAN, but when you have two WHO do not play nice with them what your options are?

    Although it would be possible to migrate the RDM using vmkfstools, I think that you will have less downtime with help e.g. robocopy. With robocopy, you can make a copy of the original, while the old ONE is still active and perform a final synchronization of data just before moving to the new LUN.

    André

  • LabManager and NetApp deduplication ratio

    We will go to the storage of one EMC to a NetApp device soon. Does anyone have figures as to how much deduplication allows, in a link to environment related clone Lab Manager? This is of course design. Does anyone have a recommendation of NetApp model for about 20 TB of usable space on disk (for LM) SAS? Overall, we expect twice the size.

    In all of our six warehouses of Lab Manager totaling about 10 TB of usable space, we see a range of 35 to 60 percent of the savings from de-duplication.  Deduplication rates are really based on how similar VM and activities you do on the same volume of NetApp (data store) which represents for various numbers see us.  On savings high volumes, we tend to have a number of very similar configurations of Lab Manager library, and from them much deployment used by our body of development, more facility similar if not identical copies of the same software in each copy.  Lots of duplicate blocks = more savings in case of deletion of duplicates.

    Data warehouses are hosted on a mixture of old fibre channel and SATA shelves and a new tray SAS.  Most of the storage is served on a three year old model FAS3040 server, only the latest SAS storage while used a FAS3160.  In our environment, two reporters are in fact a mixture of NAS services generalist as well as the storage of virtualization.  This is not ideal and caused a lot of problems (especially the storage virtualizaiton overwealming all other use...).  In general, the model FAS3160 is very well - much better than the 3040.

    One thing to consider is that the scans of deduplication, during execution, can consume a good amount of the file server CPU resources.  It is one of the biggest problems we have on the 3040 when it tries to radiation upward the two largest 2 TB FC data warehouses.   Offers of deduplication of value is exceptional, but you definitely need to ensure sufficient system capacity to deliver the savings achieved in parallel with the normal I/O to the infrastructure service attached to the spinning.

  • vWorkspace, VMware and NetApp in Multi-VLAN configuration

    Hi all.

    Today started testing the trial vWorkspace and facing a big problem with the working conditions with Netapp FlexClone and:
    Good practices place NFS traffic as a separate network and vlan, as 172.16.1.0/24 and separate a switch. Management of intellectual property, respectively, to its network and its 192168.100.0/24 vlan.

    VMmware vCenter has certainly, through ESXi, access servers, and for "vWorkspace' only watch 'NFS' network to which she has access courses. Accordingly, there is no way to specify the administration server for NetApp FlexClone.

    PS Ah, such a very small trial period and these intractable problems.

    PSPS sorry for my bad English please.

    Alexey salvation,

    vWorkspace does not take too much traffic when he speaks with a NetApp controller. You can allow the servers broker vWorkspace to access your NFS network.

    Suen

  • Masking vs. zoning for UCS and NetApp storage LUNS

    I know it's more a matter of Cisco MDS/storage, but nobody knows masking the LUNS and zoning, which one would be the most preferred method? I have two switches Cisco fabric 9148, two controllers SAN Netapp FAS3210 1 UCS chassis with 4 blades B200 M2. I said that I should not connect the Cisco UCS fabric interconnect directly at the back of the NetApp SAN and configure with LUN masking, but I rather configure a fabric zoning on switches Cisco 9148 MDS. Normally, this wouldn't be a problem but we have an offsite location where we have not all DMS switches to, and I would like to connect them directly to the San.

    I was told that this could lead to corruption if misconfigured dsik and point of view of Cisco is to use a zoning by some type of switch Cisco Fabric. Of course, I have of course this Cisco advises anyone to manage this type of configuration through their facilities instead of on the SAN. Does anyone have an opinion on the matter?

    Zoning and masking are two completely different characteristics.

    Zoning occurs on your storage switches and is the equivalent of an ACL (Access Control List).  It limits who may be considering other targets and/or initiators.  (Who I see?)

    Occurs on your storage array and masking limits what initiator LUN has access to.  (What do I see?)

    * UCS is not supported to directly connect a storage array in the interconnection of fabric, at least to have a storage upstream switch (MDS or equivalent) to push the zoning.  The use of zoning prevents a faulty initiator potentially impacting on the operation of the others by limiting what they can see in the fabric.

    If you happen to have a Nexus 5 K by chance, they can also work as your storage switch.  The N5K is able to perform almost all of the same services as MDS fabric and is fully supported.

    Kind regards

    Robert

  • Configuration of Nexus 5 k for SAN switching between UCS 6100 s and NetApp storage

    We strive to get our configuration of the UCS environment make start SAN LUN on our NetApp storage. We have a pair of Nexus 5Ks with the company / SAN license on them and the 6-port FC module. What to do to configure Nexus 5000 to operate as the SAN switch between the target storage and environment of the UCS?

    You see the 6120 WWPN on the N5Ks? What is the result of 'show flogi database' and 'show int br '.

    You have mapped the FC ports on the 6120 for the proper VSAN?

    NPIV is enabled?

  • SRM for vSphere 4.0 and NetApp

    Someone there SRM implemented in a NetApp environment ESX 4.0?

    If so, how it works?

    Here are some documents that might be useful for installing vSphere 4 and SRM with NetApp:

    NetApp vSphere SRM recommended

    http://www.NetApp.com/us/communities/tech-onTap/SRM-0708.html

    http://blog.scottlowe.org/2008/12/11/3183-VMware-SRM-on-NetApp-storage/

    =========================================================================

    William Lam

    VMware vExpert 2009

    Scripts for VMware ESX/ESXi and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

    repository scripts vGhetto

    VMware Code Central - Scripts/code samples for developers and administrators

    150 VMware developer

    Introduction to the vMA (tips/tricks)

    Getting started with vSphere SDK for Perl

    Twitter: @lamw

    If you find this information useful, please give points to "correct" or "useful".

  • Exchange 2007 and NetApp FC

    Here's the skinny:

    1200 + users. approx. 600 GB data messaging; 1 CASE/HTX server & 1 Server MBX under EXC 2007 Enterprise 64-bit, server 08.  We have a lot of equipment and storage for the design.  I have questions on is the configfor of the vm to the server of the mailbox.  I had intended to give the vm vHDD 2.  1 vmfs containing BONES and such and the 2nd vHDD would be a RDM to FC SAN volume.

    All thoughts, recommendations would be greatly appreciated.

    I would create readers separate newspapers and mailboxes, databases between 500 GB and more boxes to the letters and 100 GB for logs and each environment varies so plan accordingly.  Did you check NetApp & Exchange 2007 on VMware best practices for different scenarios and designs.  Here's a quick look at http://www-download.netapp.com/edm/TT/docs/NetApp_VMware_Exchange_TR.pdf I'm assuming you use CCR so plan carefully for optimal performance and layoffs.

    If you found this information useful, please consider awarding points to 'Correct' or 'useful '. Thank you!!!

    Kind regards

    Stefan Nguyen

    VMware vExpert 2009

    iGeek Systems Inc.

    VMware, Citrix, Microsoft Consultant

  • Alignment of disc - Windows 2003 Server and NetApp

    I recently installed a couple of the virtual machine that will host Lotus Domino providing mail to about 400 users per server.

    Virtual machines on ESX 3.5 HP DL585 4 * Quad Core with 32 GB memory, 2 * 6 HBA * NIC with a drive of 15K FC NetApp 3140c storage

    I created a VMDK file for Notes data and used a PART of DRIVE to create the partition sho it is aligned on 64 k blocks and then formatted the partition with blocks of 32 KB (in line with the alignment of disc).

    When I look in disk management, the partition created by using the PART of the DISC is a base and not on a dynamic Partition partition.

    My qustion is... If I change the partition to dynamic it cancel alignment jobs and produce a slower reaction of the I/O?

    Thank you

    Expansion VMDK and getting Windows to use space extra is also really easy too and can be done while the virtual machine is online (except the system drive, but it's just a little more effort)

    I would recommend avoiding dynamic disks. Where I work, we have thousands of Windows servers and I never remember coming across with dynamic disks.

  • Cisco, Vmware and Netapp on Netapp Flexpod document?

    Hello

    I was told during a presentation of a few months before there is a 300-page document detailing how to set up a Netapp Flexpod solution. I've looked everywhere for this document. I found a lot, for example Cisco's UCS, but not a huge detailed Deployment Guide. Can someone provide me with a link to this document, please?

    Hello

    I'm not aware of any document of 300 + page, but you could use the following as a starting point:

    Area of design for FlexPod

    http://www.Cisco.com/en/us/solutions/ns340/ns414/ns742/ns743/ns1050/landing_flexpod.html

    FlexPod deployment guide

    http://www.Cisco.com/en/us/docs/solutions/Enterprise/Data_Center/virtualization/flexpod_deploy.html

    FlexPod VMware Deployment Guide

    http://www.Cisco.com/en/us/docs/solutions/Enterprise/Data_Center/virtualization/flexpod_vsphere.html

    FlexPod for VMware deployment model

    http://www.Cisco.com/en/us/docs/solutions/Enterprise/Data_Center/virtualization/flexpod_vmware.html

    I hope this helps.

    Thank you

    Michael

  • ID of the LUN duplicate when you use several bays

    I'll put up a new battery of VMware in a large company, where EMC and Netapp without are used.

    I asked a storage engineer to present a RDM on the Netapp SAN LUN. They have gave it a LUN ID 0. Unfortunately, my boot LUN on the EMC also has a LUN ID 0. This resulted in two LUNS tables separated with the ID 0. The targets are different and the paths are all.

    I've never had duplicate LUN like this before and I was wondering if there is a problem, and if it is commonly encountered in many situations of table. Maybe I need to be more specific with SAN engineers whose LUN ID they present at the same time, is this something the SAN engineer chooses their console?

    Do I need to worry about this at all? I screen dumped two LUNs.

    I can tell you that we have duplicated LUN that come from different tables in our environment as well (that I inherited from the previous engineer).  From what I remember the pain only treats that LUN comes easily what follow up dashboard, we have no technical/performance problem that arises from this.

    I make it a point to cut LUN with different identifiers now simply for ease of tracking.

    Hope that helps a little, and someone another chimes in!

  • EagerZeroedThink! Why read i/o on the data store while inflation

    Hi all

    I got a call from a customer of 2 weeks ago asking why there reading activities on its data store when it swells a vmdk. I didn't either, and did in my Infrastructure to justify.

    Whenever I created a vmdk for EagerZeroedThick, with the Virtual Machine Wizzard (functions of Clustering Support etc.) create or inflate the context. Prezeroeing starts at 1/3 of the activity is read from the data store and the other 2/3 is in writing. I expect 100% writing to the data store since I thought that prezero a full vmdk involves only write operations (write zeros). The data store does not support vAAI and tried since with EMC and Netapp subsystems, Fibre Channel and iSCSI. Its always the same. With EMC its even more read than write operations.

    Anyone have an explanation for the behavior of ths. I looked around and didn't find anything.

    Thank you

    Harry Mitschele

    VCI and senior Consultant

    The way I understand it, the process of resetting (no matter if it has run for a newly created or an existing vmdk) zeros only blocks that are not already reset to zero. That's why he should read each block before eventually writing on it.

    Take a look at KB http://kb.vmware.com/kb/1011170 (section "vmkfstools-t0") which deals with the topic of zero setting.

    André

  • How to align data disks VMs ESXi

    Is there a way to align the ESXi VMs data (not C) disks that already have data on them without before you wipe them? A kind of ' align live. "

    If you have a NetAPP file server, you can use their tool: mbralign.

    Check if your provider of storage offers a better document practice about done (at least EMC and NetAPP) alignment.

    Read this document, it contains instructions on how to get an alignment for Windows and Linux to its customers: http://www.vmware.com/pdf/esx3_partition_align.pdf

    Microsoft offers several document and a few tools. Go to their site and search for "align the partition.

    AWo

    VCP 3 & 4

    Author @ vmwire.net

    \[:o]===\[o:]

    = You want to have this ad as a ringtone on your mobile phone? =

    = Send 'Assignment' to 911 for only $999999,99! =

  • Analyzers of vector signals OR, in real time of tektronix and tests EMC spectrum analyzers

    Normal
    0

    21

    fake
    fake
    fake

    PT - BR
    X NONE
    X NONE

    MicrosoftInternetExplorer4

    / * Style definitions * /.
    table. MsoNormalTable
    {mso-style-name: "Table normal";}
    MSO-knew-rowband-size: 0;
    MSO-knew-colband-size: 0;
    MSO-style - noshow:yes;
    MSO-style-priority: 99;
    MSO-style - qformat:yes;
    "mso-style-parent:" ";" "
    MSO-padding-alt: 0 cm 0 cm 5.4pt 5.4pt;
    MSO-para-margin-top: 0 cm;
    MSO-para-margin-right: 0 cm;
    MSO-para-margin-bottom: 10.0pt;
    MSO-para-margin-left: 0 cm;
    line-height: 115%;
    MSO-pagination: widow-orphan;
    font-size: 11.0pt;
    font family: 'Calibri', 'sans-serif ';
    MSO-ascii-font-family: Calibri;
    MSO-ascii-theme-make: minor-latin;
    MSO-hansi-font-family: Calibri;
    MSO-hansi-theme-make: minor-latin;
    mso-fareast-language: EN-US ;}

    1. how to work if vector performance of or
    Analyzers of signals compare to Tektronix real-time spectrum analyzers?

    2 can you emulate Tektronix FFT
    processing overlapping?

    3. is it possible to use vector of NOR
    Analyzers of signals of compliance EMC and/or test preconformite? Is there some
    companies use it successfully? Need a special or custom software?

    Thank you

    Hi emc2006

    I'll answer your questions separated by your topics:

    1 - What is the factor that you want to compare between these two products? In the link below, you will find the performance of the NI PXI-5660 RF Signal Analyzer system.
    2. you can develop this feature of programming in software Application development, i.e. of LabVIEW.
    3. Yes, NI´s vector signal Analyzer could run preconformite or EMC compliance analyses. In the same link below, you will find in the subdivision of Applications.

    http://zone.NI.com/DevZone/CDA/tut/p/ID/4298

    Concerning

    Napoleao
    Application engineering
    National Instruments

Maybe you are looking for