PS6000 convert RAID5 in RAID10

Similar Questions

• 2 x RAID5 against 1 x RAID10

  In my system, I can anyway have two RAID5 arrays or an array of RAID10. Now, my questions are:

  1. What is the best option for a VMware ESXi environment?

  2. what gives me better performance? I mean I understand that written RAID10 is faster than RAID5, but is this statement remains true if I write against two different RAID5 arrays?

  3. What about data security? I can see there is no difference between RAID5 and RAID10. Is this correct?

  Thank you

  Jens

  So I still wonder: are two RAID5 faster than a RAID5?

  Probably, but only marginally and only IF you have 2 separate controllers.  If you put 2 RAID 5 on the controller, which means that a single controller control groups RAID TIMES.  And for that, it would be slower, if you can stick the 2 RAID controllers to each manager RAID 5 separately, it could be a little faster, but I doubt it.  2 RAID 5 means that EACH table is still a RAID 5, RAID 10 is the fastest you can get, (other than RAID 0) period.  This requires that the same number of pins.

  RAID 5 or 6 pins, 2 with 3 axes of each or a RAID 5 RAID 10 on 6 axes.  RAID 10 beats RAID 5, RAID 50.

• How can I configure raid 1?

  I have a HP m9340f, and the 750 GB seagate drive that accompanies it is eventually died. Of course, I don't have any image or to clone, so now I have to re - install all my nonsense.

  In diagnosing the drive, I discovered that the bios has raid capability. In fact, that's what the reader has been set on (although there was still only 1 disc). I've never used raid, so my question is, if I want to configure raid 1, so that the next time a drive dies, I have a built-in backup, what exactly do I do? I know that I need to install 2 hard drives, but as soon as I did this, raid configure automatically? I don't see any special properties in the screens of the bios for raid, except for access to the drive (ide/raid/AHCI). Would this be a good idea to use seagate disks, or cares not the bios?

  Any help would be appreciated.

  Hello bassman952, maybe I can offer a glimpse of the Raid arrays. Many Members here know this already, but for those who do not, it seems that your system is capable of Raid. It has the Intel G33 chipset, where specifications for this chipset can be found here. This chipset includes the Raid feature for most Raid array configurations. However, HP changed the BIOS to only supported Bay Raid0 and Raid1. A Raid0 array consist usually of 2 hard drives and recommend that be of the same manufacturer, model, capacity, speed and cache. It is possible to use different brands and other specifications of the hard drive, but it is strongly discouraged. A Raid1 array also usually consist of 2 hard drives with the same configuration as described above. All Raid arrays must be installed before all the data is stored on the table.  To configure a matrix Raid on the chipset Intel, immediately after the HP Logo screen disappears, press ctrl and the I touch both and the Intel Matrix Raid Array utility will appear. You can then configure matrix Raid type and disk to be included in this table, save the configuration and exit. The table written Raid0 and bed for two drives at the same time, but with different data. This is referred to as a matrix of blocks. The Raid1 array is referred to as a network of mirrors. If one of the disks in Raid0 array were to fail, all data on both drives would be lost. The Raid1 array reads and writes the same data on two drives in the Raid array at the same time. If one of the disks were to fail, a new hard drive can be installed to replace the failed drive and the system booted and the Raid1 array will be rebuilt, which means that the data from the hard drive of existing work will be copied to the new hard disk. Depending on the capacity of the hard disk and the speed of the system, it can take a good amount of time to accomplish. After that the Raid1 array is rebuilt, the system should boot normally. Table of Raid0 is mainly used for performance while the Raid1 array is used mainly as a redundant to the first hard drive backup. If a Raid array is broken, which means that the configuration of the table is changed, all data on the hard disks will be lost, even in the Raid1 array. Can I have left some items, since I'm trying to remember this memory, but this is essentially how HP uses Raid arrays in their use of home systems. Others use Raid arrays that are different, such as Raid5 and Raid10, Raid0 + 1 and some other Raid configurations, but since we are talking about the Raid arrays on HP systems, other table Raid differences would be not useful for the discussion here.

• New installation of 6.1 of the Horizon. It is normal that provisioning linked clones will take a lot of time?

  Hello

  First, I will describe my setup:

  1 Management Server: Dell PowerEdge R730 48 cores, 2.5 Ghz with 9 drive HARD 300 GB SAS 15 K (RAID 5) and 256 GB of RAM

  2 hosts VDI server: Dell PowerEdge R730 48 cores, 2.5 Ghz with 9 drive HARD 300 GB SAS 15 K (RAID 5) + 1 SSD 200 GB, 256 GB of RAM, 2 K1 grid cards

  1 unit number logic on a Dell Equallogic PS6110X (750 GB) where I put the masters.

  1 switch Dell N4032F (10 GB)

  All the server has 4 10 GB ports configured with NPAR (so, I have virtual network card of 6 x 10 GB for storage and 6 virtual network adapter for the network of the VM) and 2-port 1 GB for the management.

  My master: Windows 7 SP1 x 86 optimized with 24 GB HDD Thin-Provisionned (2 vCPU 2 GB of RAM, 3 VMXNET network Card). There is no application has added to this master.

  I can't ask the Microsoft VMXNET 3 fix because it is said that this patch is not applicable.

  I have no VSAN because that only 2 VDI hosts, so I set up my stock for each guest with local storage for the linked clones and SSD local for the replica.

  When I run the provisioning of the office pool (17 VM):

  The clone of the replica (Equallogic == > local SSD) aura 2'30 "... It's good for me (113 MB / s... the Equallogic is also used for another another LUN storage server)

  But the rest of the entire operation completes the provisioning will take 40'.

  More than 50% of carrots is around 100% of the CPU on the host.

  I tried to change my (9 disc hard + 1 spare) RAID5 to RAID10 (10 HDD). It will take 50'.

  And, when all of the operations are completed. I have 2 or 3 VM with the message: Agent unreachable and 1 or 2 crashes on the provisioning.

  One recompose your desktop one VM will take about 1'30.

  My question: it is normal that commissioning will take so much time?

  And another question: when I'm connected to a virtual desktop computer with customer Horizon during a pool commissioning, the freezing of the VM or I can't just plug. Although I Office ready. And sometimes, vCenter client will freeze to a large number of seconds during deployment

  Have someone idea?

  PS: Sorry for my approximate English ;-)

  Solution: I replaced the lsi_mr3 of storage vmware with the driver megaraid_perc9 driver.

  Thanks to ddewire for his help.

• Hard disk RAID configuration for R610

  Hi all

  I'm new to virtualization, and after reading the documents a lot, a lot, I think I'm starting to get my head around the concepts in the installation and operation of ESXi and VMs.

  My plans are to purchase a Dell R610 with the 'ESXi v4.0 2CPU Embedded, 0 year Upgrade Subscription basis, on SD card' which I think gives me basic ESXi host that can be managed remotely using VMWares free Vsphere Client?

  I'm not sure however how best to configure the machine hard drives.

  Initially, I was counting RAID1 on 2 x 15 k 146 GB and RAID5 (1 HS) on 4 x 10 k 146 GB with the RAID1 for readers of virtual machine system and RAID5 for data etc.

  But I think I can miss something fundemental of how VMs are different to normal physical machines? Is it easy/possible/useful to have a VMs data on a separate hard drive to his BONES? None of the VMS provided for are very taxing on the material.

  Is it better to say just to have the (1 HS) 5 x 146 GB RAID5 or RAID10 6 x 146 GB 15 k vs. 10 k would make much difference?

  Currently all virtual machines and their data will be stored locally. The delve into virtualization to be successul then focus on DAS or more probably a SAN iSCSI in the future.

  My apologies that it is largely on the material possible, rather than just on ESXi. I hope that you can be able to give some advice.

  Generally agree with the sixth, what is needed is a storage subsystem able to provide the number of the IOPS / s random evenly as required by multiple virtual machines running.

  The PERC 6i in these machines is a controller fast - so a RAID-5 volume with many drives high speed as possible give the best performance. Don't bother separating out the VM disks, it was never really advantage anyway, and waste IOPS / s because there are less pins available to deal with the simultaneous overall workload.

  Do not forget to spec the BBWC (think it's standard on this server in any case).

  Also, by the way, I noticed that the cost is almost cut in half with 48GB of RAM as apposed to 64 GB.

  Please give points for any helpful answer.

• 1 TB SAS 7.2 K vs 1 TB SATA 7.2 K (DELL MD3000i)

  Hi guys,.

  I'm new to the SAN thing and hope someone can enlighten us.

  I'm playing between the use of drives of 1 TB SAS or SATA 1 TB for my new DELL MD3000i SAN. Both drives work on 7.2 K rpm. It will be connected to the ESX and run "normal" virtual machines such as files and print Windows, Exchange, SQL, Linux, etc. Nothing very heavy.

  A. you have any input on how the performance, reliability, etc. would be like between these two types of drives - SAS 7.2 K vs 7.2 K SATA?

  (B) with 14 x 1 TB 7.2 K drives, should I run RAID5 or RAID10 to maximize performance? IIRC, 14 disks in a RAID5 configuration will have a rate of 14 pins, while 14 disks in RAID10 will throughout the 7 pins. Capacity is not a problem right now.

  (C) on a note related, how about 14 x 450 GB SAS 15 K disks? Their execution in RAID10 is a complete overkill? Or RAID5 would be the sweet spot since the disks are already 15 K RPM?

  Thank you very much!

  My understanding is that the SAS drive interfaced will be 10-15% faster for the random i/o, even if speed control 7.2 k is the same. It is the ability of the difference between the fairly simple, integrated SATA on the drive controller and the SAS controller with double vent more advanced. So, all things being equal elsewhere the SAS interfaced by car will be a better choice for general use. For purposes of archiving, for example, there would be little difference. Pointing to the MTBF numbers are similar, readers are probably the same as their controller.

  What levels of raid are in my opinion is one of the determinants for raid 10 is the problems that can occur with a raid 5 is the rebuild time and performance hit.

  If you have several virtual machines running in production and you have a raid 5 rebuild during a working day, is it acceptable to have a lot of your virtual machine runs very slowly.

  If it is not then use raid 10 and get the best performance as well.

• Storage VMFS Partition Performance

  I am trying to determine if I'll get best performance of 6-300 Gb 15 K FC disks in a table 6 FC - 300 GB 15 K in a table of RAID5 or RAID10?

  I don't have much time to really sit and Experiment (would like to but time = $$).

  This is the table that will store between 5-9 WIN2k3 VM.

  I like the idea of RAID 10, which allows more than 1 drive fail (as long as it isn't the same mirror) and still survive with times of rapid reconstruction.

  This question has been turned off before, but not with the amount of readers, I want to say

  Any help is appreciated!

  If you have a load of joint task force (most do) then RAID10 will give good performance than RAID5. RAID5 has faster sequential readings as RAID10, but RAID10 is normally the one who gives you the best performance in a VMware environment where multiple virtual machines are access to the same storage.

  Lars

• ASM external redundancy and the number of disks

  Hello
  
  
  We are about to install a solution RAC on IBM AIX 5.3 LPAR and VIO.
  According to Oracle certification, we are forced to use DSO (11 g) for storage.
  Because we have a huge SAN solution for all our systems, also the installation of the DSO will run on the SAN
  The complete SAN is in a RAID5 or RAID10 configuration. This automatically means that we will use external redundancy
  The database that will be placed on the system is currently around 50 GB in size.
  
  Now that you guys and women think how to set start, if
  
  one)
  We set up initially only 1 disc in the diskgroup to 60 GB for example,
  ASM does support the ability to increase a logical disk, because we are currently used, using the VIO Server?
  If it is not supported there so if the diskgroup needs to be extended, add another disk the same size, or we can add disks with smaller sizes
  
  (b)
  We set up initially more discs of small size in the diskgroup, making a total of 60 GB
  ASM will probably begin the distribution of the data on the three disks.
  Overhead that does?
  That this has a negative impact, because the three logical drives are already in a striped solution (RAID10)?
  
  
  (If the extension of the logical drive using the VIO in Setup a) only supported, automatically end up in b).
  But I would like to know head/experiences and advice.
  
  Thanks in advance
  
  FJF

  You can increase the size of the logical unit number, which is a Member
  ASM diskgroup, after that, to increase
  the disk partition of this LUN to cover the new size of this LUN,
  and the last step is to tell ASM this member disk has been resized
  "alter diskgroup resizing disk.

• Online support for RAID5-10 migration on H700

  Our client has a T610 with a H700 integrated controller functioning as a host VMware 5.5 server there currently 6 disks in a single Bay RAID5.  The current RAID5 LUN consumes all available space, although the actual allocated storage is much less.

  Let's change the function of the server significantly, adding an app of the important data.  As part of this, I would like to add a couple more readers and convert the table to a pair of paintings RAID10 4 discs.

  I don't want to go into details of the allocated space, the size of the LUN, etc, here - we can manage that.  My question is, if the allocated space corresponds to the size of the target table I can make this migration online?

  Key questions:

  -I can reduce the size of the RAID5 LUN at all or anywhere near actual allocated space?

  -can I convert table RAID5 RAID10 - once again, assuming that allocated space should be?

  -I'd better first add additional disks and converted to a table of RAID10 8 - drive and then divide into two sets of 4 disks?

  -What specific Dell server management software must be installed?   Or is it only possible at the start of the BIOS?

  Of course, I make backups first!

  -Ken

  Regarding the managed material VD (tables/MON) go to:

  You can never "shrink" a picture by reducing the number of discs or the allocated space.

  Controllers do not support migration or nested RAID levels (10/50/50)... for example, you cannot add two drives to 4-drive RAID 10 either.

  If you plan to switch to a RAID 10 (which I support), then you will have to the data being backed up and restore in the new array.

• RD350 configure RAID10 disk

  We have RD350 with 4xSAS 2 TB HDD, RAID500 (LSI SAS 9240-8i) card and I can't set it up 4 disks as RAID10 with full capacity.

  I tried capacity 1500 GB and also it is not.

  My practice (with the last FW - BIOS, RAID, TSM, PSU)

  The BIOS > Startup Manager > adapters and UEFI > LSI MegaRAID... > configuration management > create Virtual Drive.

  Select RAID10, select all 4 disks good unconfigured (in Bay 1) and after save configuration it wil display:

  To create virtual drives for the selected RAID level, more lasting and an equal number of readers in each span is needed. Click OK to go back and change.

  WHY?

  I have only 1 Bay and 4 disks selected in 1 Span.

  (RAID5 is not accessible)

  Help, please!

  TNX

  mirekh

  Try to create the first Bay with only 2 drives and then create a different scale with the other 2 drives.

• migration of RAID5 at 50

  Hello

  We currently have a single PS4000 with 8 disks in RAID-5.

  We plan to add another 8 disks and convert the SAN RAID-50...

  Are we able to add the extra 8 disks and create a new pool of storage with RAID-50? We can then migrate existing virtual machines across to the new pool.

  Once all over the place, we plan to blow the Bay RAID5 and add disks to the RAID-50.

  Is this possible?

  See you soon,.

  H

  Hello

  A member cannot be in two pools at the same time.   After attaching a RAID level, when you add disks, they are added to the RAIDset existing at the same level.

  Sorry, but you can't be hidden from a larger to a smaller RAID level.  R50-> R5 Yes, but not R5-> R50.

  You will have either to back up data, reset and restore, or get an array of loan, add it to the group, delete the original table, and then add back to R50.  The removal of the loan.   This can be done directly without any loss of data or downtime.

  Kind regards

  Don

• PERC S300 RAID10 Drive upgrade Question

  I currently have 4x250GB HARD drives in a PERC S300 on a PE T110.  It works fine but I really want to upgrade to SSD because they are becoming so cheap.  However, when I was in RAID5, it requires that any replacement of disks are of equal size or greater, even if the actual volume was a fraction of the available space.  Is this the same requirement on the S300 for RAID10?

  Yes, it's necessary. In addition, you cannot mix SAS and SATA and SSD, SATA or SAS, SSD, so you will need to do a backup/restore of your SSD barrette.

• Oracle 11g DW - RAID5 or RAID1 + 0?

  Hi all
  
  For an implementation of DWH Oracle 11 g, what drive arrangement is preferable, if only load & the data query performance are taken into account: RAID1 + 0 or RAID5?
  
  In addition, only RAID5 also automatically given stripe (i.e. RAID5 = RAID5 + 0)?
  
  Ofcourse RAID5 will protect only against a single disk failure (in the case of no help). with RAID1 + 0 protection against multiple disk failures.
  
  Please tell us what is the preferred option.

  OracleDisciple wrote:
  As this would be an implementation of the warehouse with loading data pure data per day during off-peak hours and sessions of massive request for the rest of the day performed once, RAID5 give equal or slightly lower performance compared to RAID1?

  Unknown. How the storage system implements RAID determines productivity. RAID5 can be indeed not faster than RAID10 if RAID5 is made for a better storage system architected and RAID10.

  The coefficient of basis (con), making it the most expensive RAID5 is calculation of parity. What makes cheaper RAID10 (pro) is to have a disk mirror set (and a whole bunch of disk controllers extra) to maintain the e/s readings.

  Modern storage system, there are two features that often deny these pros and cons parity calculations are discharged to the storage group. So write an I/O for RAID5 is fast for RAID10 as the calculation of parity does not affect the waiting time (from the database for example) for that i/o to complete.

  The second feature is the in-memory cache. Modern storage systems may have caches of 0.5 to RAM and bigger memory - and the memory are inexpensive, large storage system caches become the norm. This means that the vast majority of all i/o on the storage system may be against the cache memory - so little time spend waiting for disks to towers.

  Recommendation of Oracle is SAMI - Stripe and mirror everything. This is mentioned in the Metalink, books of kitchen CARS and best practices notes and so forth. At the same time you can get a white paper from EMC for example, co-written by Oracle Corp., proclaiming just how fast and efficient RAID5 on EMC storage array networks are.

  So it is more clear and decisive line between different implementations of RAID. We must look at just how the storage array is built and well he is capable of dealing with a specific implementation of RAID.

• So with the new update every attachment I get an Outlook user is converted in a winmail.dat file.  How Apple could make a such big screw-up?  Even when I check gmail via apple mail conversion is made, but if I connect to google, the files are

  Why is each attachment outlook, I get in the Mail now converted in a winmail.dat file?  I don't want to go make an app to allow me to open these files - it is something stupid on the part of Apple - please correct!

  Tell Apple - http://www.apple.com/feedback/

• Convert AAC to MP3 Itunes - 10 Windows files

  ITunes Version 12.5.1

  Windows 10

  I would like to convert AAC to MP3 audio files.  How can it be done?  I would like to transfer my songs on a flash drive to plug into my car audio system to play my songs.  The interesting thing is that on my drive hard it shows that the music file is MPEG-4 audio type - this type of file will play on stereo car if I copy these files directly to a flash drive.

  There was a time when we could select all the songs, right click and select Convert to MP3.

  Thank you, in advance for your answer.

  (NOTE: the songs are already on my drive hard and do not intend to import them again)

  (NOTE 2: If someone used AnyTans - for Windows?) If so, what are your thoughts on his performance?)

  MPEG-4 audio is another description for files to AAC format or .m4a.

  Import or convert the format is defined in Edition > Preferences > General > import settings. You will lose a little quality by converting, so you would normally keep the originals. Define a flow similar to the original. Don't serve him no 128 k AAC 320 k MP3 conversion.

  TT2