1 2 Previous Next 27 Replies Latest reply: Mar 3, 2014 8:41 AM by daveonthemac Go to original post
  • 15. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    Smphoto74 Level 1 Level 1 (5 points)

    I have my R4 unit set in Raid 10.  Speed test on the R4 are 220/200 write/read however copying files to and from internal 1TB HD on a new iMac maxes out at 130MBps which kinda *****....only twice faster than my FireWire HD.  My data is super important but I would like to do Raid 0 and just rely on my Time Machine Backup.  I have thought about putting a TM backup in Raid 1 so that I technically have 2 back ups but on a FW HD instead of using the R4 in any sort of backup mode.  Would you recommend that when data is super sensitive? 

     

    I wish someone would post some real time copy speeds instead of speed test results.  With my R4 unit 3/4 full (1.5TB out of 2TB, raid 10) I am getting about 150MBPS on the speed test but still a good 120MBps on copy and paste so there is some sort of bottle neck there.  I'm guessing the internal HD.  Would a Internal SSD improve the copy and paste to match the speed test?  I just used the free blackmagic speed test from the App store.

  • 16. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    RatVega™ Level 4 Level 4 (1,910 points)

    IMHO, RAID 10 doesn't make the sense that RAID 5 does (but it's easier to implement.)  What I especially don't like is losing half the storage...  I ran RAID 0 (still do on one system) for data that really needs to be fast (e.g., uncompressed HD video) but only because it was all I could afford. I built a 2TB RAID 0 box for $1200 when the asking price for a 2TB hardware RAID 5 was $8-10K.

     

    Back-up is important no matter what technology you use, but (again, IMHO) Time Machine isn't really meant for the job since it wants to "track changes in your documents" and at least the way I work, I can make 50-100GB of "changes" in a hard day of editing. That's going to create a problem on a 1TB or 2TB TM box...

     

    The thing about very fast disk arrays is that they're hard to measure in the real world because to see the crazy fast stuff you need to have really fast arrays at both ends and a clean path in between. I had exactly that a while back when I had two RAIDs attached to the same system and was backing up a feature film I had just finished cutting. It took about 22 minutes to back up something 318GB of source and work files. If I'd been going to a clean single SATA drive it probably would have taken on hour.

     

    Solid State Drives are a really interesting item. They are terrifyingly fast but pretty small and quite expensive. Yes, they should improve your cut/paste scores, but how fast can you afford to go?  The scenario that I like best for SSD is to use it as a system drive with a reasonably high (4-6) spindle RAID5 as main storage. In this scenario on my system (8X 3.2GHz Xeon) it may be possible to get 250MB/sec throughput. Some of the things I do (like transcoding HD video) can currently saturate the 8 processors and become disc-dependent for performance.

     

    But the thing to keep in mind is that in the end, when you raise the bar you're just moving the choke point to a more expensive place. 

  • 17. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    Smphoto74 Level 1 Level 1 (5 points)

    RatVega, 

     

    What do you recommend using for backup software since you don't think TM is up to it?  I'm a photographer an I have 3 employees editing images all from the r4 unit at any given time.  I have it TM to a WD 2TB HD(USB) and really haven't had any issues. 

     

    I believe and I would have to double check but in Raid 10, the write is better than in Raid 5.  I do a ton of writing so that is why I went Raid 10.  I really want Raid 0.  Lol.

  • 18. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    RatVega™ Level 4 Level 4 (1,910 points)

    If you like TM then stick with it.  I tend to think in video terms and (as I mentioned) I can see myself wiping out multi-TB drives at a pretty good clip. In your case, you don't have the render files and versioning issues so a couple of TB will last you a great deal longer.

     

    By all means, do what YOU think is best for your circumstance. I'm just a guy (in a different business) with a little time on the ground and an opinion. What I wanted to give you is an alternative that I think makes sense.

     

    Speed is a relative thing. Once you are going faster than you need to, you start looking at different things. My recommendation for a SATA III RAID 5 is based on a couple of points:

    1. SATA III is twice the speed of SATA II (which is basic to a Mac Pro) and this speed costs you the price of the RAID controller.

    2. With twice the theoretical speed, staying above your desired threshold should be possible without a lot of tuning.

    3. Since a RAID 10 is mirrored, your net capacity is half, a fairly steep price when RAID 5 provides 75% in a 4-spindle array, more if you go with more spindles.

    4. I assumed you were using RAID 10 because that is what was offered (as I said, it's easier.) This is apparently not true and your opinion of what you need trumps my opinion any time. 

     

    I've had the RAID 0 and I understand the draw...  If you're not living on the edge, you're taking up too much room!    (but it's a lot like playing Russian Roulette with a .45 Automatic...)

  • 19. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    Smphoto74 Level 1 Level 1 (5 points)

    Different views are great and a reason why I'm here so thanks for your comments and tips.

     

    I mainly did Raid 10 due to lots of research online about what Raid to use for whatever.  Having only 2TB available out of 4TB is fine with me so that's never been an issue.  Speed is important...it's a guy thing I think. Lol

     

    I'm seriously thinking of doing a Raid 0 in my R4 unit then doing TM in Raid 1 on my 4TB WD FireWire drive.  Granted that is only 2 TB for TM but I never plan to take the R4 over 2TB.   I rotate my work out fast and I've always heard drive performance starts to bog down after 1/2 full.  I do also keep another copy of the original files in a separate location.

     

    I did buy FCPX and want to get into video a bit. More personal stuff than business.  Lol.

     

    Thanks again for your help 

  • 20. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    Smphoto74 Level 1 Level 1 (5 points)

    RatVega™ wrote:

     

    IMHO, RAID 10 doesn't make the sense that RAID 5 does (but it's easier to implement.)  What I especially don't like is losing half the storage...  I ran RAID 0 (still do on one system) for data that really needs to be fast (e.g., uncompressed HD video) but only because it was all I could afford. I built a 2TB RAID 0 box for $1200 when the asking price for a 2TB hardware RAID 5 was $8-10K.

     

    Back-up is important no matter what technology you use, but (again, IMHO) Time Machine isn't really meant for the job since it wants to "track changes in your documents" and at least the way I work, I can make 50-100GB of "changes" in a hard day of editing. That's going to create a problem on a 1TB or 2TB TM box...

     

    The thing about very fast disk arrays is that they're hard to measure in the real world because to see the crazy fast stuff you need to have really fast arrays at both ends and a clean path in between. I had exactly that a while back when I had two RAIDs attached to the same system and was backing up a feature film I had just finished cutting. It took about 22 minutes to back up something 318GB of source and work files. If I'd been going to a clean single SATA drive it probably would have taken on hour.

     

    Solid State Drives are a really interesting item. They are terrifyingly fast but pretty small and quite expensive. Yes, they should improve your cut/paste scores, but how fast can you afford to go?  The scenario that I like best for SSD is to use it as a system drive with a reasonably high (4-6) spindle RAID5 as main storage. In this scenario on my system (8X 3.2GHz Xeon) it may be possible to get 250MB/sec throughput. Some of the things I do (like transcoding HD video) can currently saturate the 8 processors and become disc-dependent for performance.

     

    But the thing to keep in mind is that in the end, when you raise the bar you're just moving the choke point to a more expensive place. 

    I been looking for it and finally found it....this is what I read online and what I based my decision on going to Raid 10 instead of 5.   It's a long read. lol

     

     

    RAID5 versus RAID10 (or even RAID3 or RAID4)

     

    First let's get on the same page so we're all talking about apples.

     

    What is RAID5?

     

    OK here is the deal, RAID5 uses ONLY ONE parity drive per stripe and many

    RAID5 arrays are 5 (if your counts are different adjust the calculations

    appropriately) drives (4 data and 1 parity though it is not a single drive

    that is holding all of the parity as in RAID 3 & 4 but read on). If you

    have 10 drives or say 20GB each for 200GB RAID5 will use 20% for parity

    (assuming you set it up as two 5 drive arrays) so you will have 160GB of

    storage.  Now since RAID10, like mirroring (RAID1), uses 1 (or more) mirror

    drive for each primary drive you are using 50% for redundancy so to get the

    same 160GB of storage you will need 8 pairs or 16 - 20GB drives, which is

    why RAID5 is so popular.  This intro is just to put things into

    perspective.

     

    RAID5 is physically a stripe set like RAID0 but with data recovery

    included.  RAID5 reserves one disk block out of each stripe block for

    parity data.  The parity block contains an error correction code which can

    correct any error in the RAID5 block, in effect it is used in combination

    with the remaining data blocks to recreate any single missing block, gone

    missing because a drive has failed.  The innovation of RAID5 over RAID3 &

    RAID4 is that the parity is distributed on a round robin basis so that

    there can be independent reading of different blocks from the several

    drives.  This is why RAID5 became more popular than RAID3 & RAID4 which

    must sychronously read the same block from all drives together.  So, if

    Drive2 fails blocks 1,2,4,5,6 & 7 are data blocks on this drive and blocks

    3 and 8 are parity blocks on this drive.  So that means that the parity on

    Drive5 will be used to recreate the data block from Disk2 if block 1 is

    requested before a new drive replaces Drive2 or during the rebuilding of

    the new Drive2 replacement.  Likewise the parity on Drive1 will be used to

    repair block 2 and the parity on Drive3 will repair block4, etc.  For block

    2 all the data is safely on the remaining drives but during the rebuilding

    of Drive2's replacement a new parity block will be calculated from the

    block 2 data and will be written to Drive 2.

     

    Now when a disk block is read from the array the RAID software/firmware

    calculates which RAID block contains the disk block, which drive the disk

    block is on and which drive contains the parity block for that RAID block

    and reads ONLY the one data drive.  It returns the data block.  If you

    later modify the data block it recalculates the parity by subtracting the

    old block and adding in the new version then in two separate operations it

    writes the data block followed by the new parity block.  To do this it must

    first read the parity block from whichever drive contains the parity for

    that stripe block and reread the unmodified data for the updated block from

    the original drive. This read-read-write-write is known as the RAID5 write

    penalty since these two writes are sequential and synchronous the write

    system call cannot return until the reread and both writes complete, for

    safety, so writing to RAID5 is up to 50% slower than RAID0 for an array of

    the same capacity.  (Some software RAID5's avoid the re-read by keeping an

    unmodified copy of the orginal block in memory.)

     

    Now what is RAID10:

     

    RAID10 is one of the combinations of RAID1 (mirroring) and RAID0

    (striping) which are possible.  There used to be confusion about what

    RAID01 or RAID10 meant and different RAID vendors defined them

    differently.  About five years or so ago I proposed the following standard

    language which seems to have taken hold.  When N mirrored pairs are

    striped together this is called RAID10 because the mirroring (RAID1) is

    applied before striping (RAID0).  The other option is to create two stripe

    sets and mirror them one to the other, this is known as RAID01 (because

    the RAID0 is applied first).  In either a RAID01 or RAID10 system each and

    every disk block is completely duplicated on its drive's mirror.

    Performance-wise both RAID01 and RAID10 are functionally equivalent.  The

    difference comes in during recovery where RAID01 suffers from some of the

    same problems I will describe affecting RAID5 while RAID10 does not.

     

    Now if a drive in the RAID5 array dies, is removed, or is shut off data is

    returned by reading the blocks from the remaining drives and calculating

    the missing data using the parity, assuming the defunct drive is not the

    parity block drive for that RAID block.  Note that it takes 4 physical

    reads to replace the missing disk block (for a 5 drive array) for four out

    of every five disk blocks leading to a 64% performance degradation until

    the problem is discovered and a new drive can be mapped in to begin

    recovery.  Performance is degraded further during recovery because all

    drives are being actively accessed in order to rebuild the replacement

    drive (see below).

     

    If a drive in the RAID10 array dies data is returned from its mirror drive

    in a single read with only minor (6.25% on average for a 4 pair array as a

    whole) performance reduction when two non-contiguous blocks are needed from

    the damaged pair (since the two blocks cannot be read in parallel from both

    drives) and none otherwise.

     

    One begins to get an inkling of what is going on and why I dislike RAID5,

    but, as they say on late night info-mercials, there's more.

     

    What's wrong besides a bit of performance I don't know I'm missing?

     

    OK, so that brings us to the final question of the day which is: What is

    the problem with RAID5?  It does recover a failed drive right?  So writes

    are slower, I don't do enough writing to worry about it and the cache

    helps a lot also, I've got LOTS of cache!  The problem is that despite the

    improved reliability of modern drives and the improved error correction

    codes on most drives, and even despite the additional 8 bytes of error

    correction that EMC puts on every Clariion drive disk block (if you are

    lucky enough to use EMC systems), it is more than a little possible that a

    drive will become flaky and begin to return garbage.  This is known as

    partial media failure.  Now SCSI controllers reserve several hundred disk

    blocks to be remapped to replace fading sectors with unused ones, but if

    the drive is going these will not last very long and will run out and SCSI

    does NOT report correctable errors back to the OS!  Therefore you will not

    know the drive is becoming unstable until it is too late and there are no

    more replacement sectors and the drive begins to return garbage.  [Note

    that the recently popular IDE/ATA drives do not (TMK) include bad sector

    remapping in their hardware so garbage is returned that much sooner.]

    When a drive returns garbage, since RAID5 does not EVER check parity on

    read (RAID3 & RAID4 do BTW and both perform better for databases than

    RAID5 to boot) when you write the garbage sector back garbage parity will

    be calculated and your RAID5 integrity is lost!  Similarly if a drive

    fails and one of the remaining drives is flaky the replacement will be

    rebuilt with garbage also propagating the problem to two blocks instead of

    just one.

     

    Need more?  During recovery, read performance for a RAID5 array is

    degraded by as much as 80%.  Some advanced arrays let you configure the

    preference more toward recovery or toward performance.  However, doing so

    will increase recovery time and increase the likelihood of losing a second

    drive in the array before recovery completes resulting in catastrophic

    data loss.  RAID10 on the other hand will only be recovering one drive out

    of 4 or more pairs with performance ONLY of reads from the recovering pair

    degraded making the performance hit to the array overall only about 20%!

    Plus there is no parity calculation time used during recovery - it's a

    straight data copy.

     

    What about that thing about losing a second drive?  Well with RAID10 there

    is no danger unless the one mirror that is recovering also fails and

    that's 80% or more less likely than that any other drive in a RAID5 array

    will fail!  And since most multiple drive failures are caused by

    undetected manufacturing defects you can make even this possibility

    vanishingly small by making sure to mirror every drive with one from a

    different manufacturer's lot number.  ("Oh", you say, "this schenario does

    not seem likely!"  Pooh, we lost 50 drives over two weeks when a batch of

    200 IBM drives began to fail.  IBM discovered that the single lot of

    drives would have their spindle bearings freeze after so many hours of

    operation.  Fortunately due in part to RAID10 and in part to a herculean

    effort by DG techs and our own people over 2 weeks no data was lost.

    HOWEVER, one RAID5 filesystem was a total loss after a second drive failed

    during recover.  Fortunately everything was on tape.

     

    Conclusion?  For safety and performance favor RAID10 first, RAID3 second,

    RAID4 third, and RAID5 last!  The original reason for the RAID2-5 specs

    was that the high cost of disks was making RAID1, mirroring, impractical.

    That is no longer the case!  Drives are commodity priced, even the biggest

    fastest drives are cheaper in absolute dollars than drives were then and

    cost per MB is a tiny fraction of what it was.  Does RAID5 make ANY sense

    anymore?  Obviously I think not.

     

    To put things into perspective: If a drive costs $1000US (and most are far

    less expensive than that) then switching from a 4 pair RAID10 array to a 5

    drive RAID5 array will save 3 drives or $3000US.  What is the cost of

    overtime, wear and tear on the technicians, DBAs, managers, and customers

    of even a recovery scare?  What is the cost of reduced performance and

    possibly reduced customer satisfaction?  Finally what is the cost of lost

    business if data is unrecoverable?  I maintain that the drives are FAR

    cheaper!  Hence my mantra:

     

    NO RAID5!  NO RAID5!  NO RAID5!  NO RAID5!  NO RAID5!  NO RAID5!  NO RAID5!

     

    Art S. Kagel

  • 21. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    Smphoto74 Level 1 Level 1 (5 points)

    I'm putting my Pegasus R4 in Raid 0 this weekend.  I also want my OS(lion) on there as well.  Not sure if I will make a partition for that or not but my backup will be to a 3TB WD drive via TM and also another backup to another WD HD using Carbon Copy cloner set up for Daily backup.  So 2 backups should be pretty good.  Granted if a drive dies on the R4 it will still suck but either way, it would still suck if a drive went out in Raid 10. 

     

    I only have a 1TB HD in my iMac so I figure I can get better performance from my programs if they are installed on the Pegasus R4(although I heard boot times will still be same as a regular HD.  I will post speed of the R4 in Raid 0 Monday. 

  • 22. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    megalaser Level 1 Level 1 (10 points)

    So what is faster then, a disk connected to the internal SATA bus or THE SAME disk connected via Thunderbolt.

  • 23. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    The hatter Level 9 Level 9 (58,880 points)

    rather easy to find good quality backup programs, if you look even slightly. 

     

    RAID6 is worth looking at also.

  • 24. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    willp1 Level 1 Level 1 (75 points)

    Thunderbolt by wide margin.

     

     

    Will.

  • 25. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    Tman123 Level 1 Level 1 (0 points)

    Are you sure?

     

    So if we have 4 x 2TB Drives in an external Thunderbolt Case, and we have 4 x 2TB Drives esata internal (Both Western Digital Black Radi 0) something like FCPX will work faster with theThunderbolt?

  • 26. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    RatVega™ Level 4 Level 4 (1,910 points)

    You're asking the wrong question...

     

    Maybe I've missed something here, but I don't believe that Thunderbolt is available on a Mac Pro. I think I recall seeing an aftermarket card, but that would be gated by PCI speed so it would end up being about the same speed.

     

    Given enough spindles, Thunderbolt should be faster because the internal SATA bus on a Mac Pro is SATA II limited to 4 drives. OK, 6 drives on some Mac Pros...  On the other hand, T-bolt can be whatever is offered: SATA3, lots of spindles, neon lights, dancing girls...  just not in a Mac Pro. 

     

    If you really need dead fast data at very high speed for your Mac Pro, you should look at SATA3 or fiber channel. It is possible to convert the internal SATA to SATA3 but the ability to boot from the RAID is expensive. One alternative is an SSD boot drive on the internal SATA bus. I honestly know little about fiber channel except that I can't afford to go that fast.

  • 27. Re: What's faster? Built in hard drive or Promise Pegasus R4 4TB (4x1TB) RAID System connected with Thunderbolt?
    daveonthemac Level 1 Level 1 (0 points)

    http://www.tomshardware.com/reviews/thunderbolt-performance-z77a-gd80,3205-2.htm l

     

    According to this article, RAID or not, any external hard drive will suffer from an extra layer of thinking than using an internal drive.  SATA or eSATA talks the native language, so there's no translation going on before writing or collecting your data.  So, lets say you have 1 million 4 KB files you have to write, vs 1 single 4GB file, the internal drive will perform better than a RAID array with the 1 million 4 KB files.  The RAID will perform better with the 1 single 4GB file, because it only has to think about translating the data 1 time, vs 4 million times. Even though it's the same amount of data.  If you want to know the details, just check the link.  If you use a Thunderbolt drive with no RAID array, you have a chance of doing much closer to the performance of the internal drive, depending on the drive and what controller it uses.  That's mentioned towards the bottom of the article.

1 2 Previous Next