Resolved SSD raid 0 w/ SATA II and SATA III

Discussion in 'MacBook Pro' started by GenderFi, Dec 10, 2012.

  1. GenderFi, Dec 10, 2012
    Last edited: Dec 11, 2012

    GenderFi macrumors regular

    Joined:
    Sep 18, 2012
    #1
    I have an early 17" MacBook Pro 2011 model which has SATA III main bay and SATA II optical bay. I would like to setup them up with SSD's as raid 0 (my first raid setup). From what I understand going raid 0 is basically combining them as one hard drive doulbing the speed (i.e. 500MB/s + 500MB/s = 1000MB/s). Would it be possible to mix and match? Would I be able to use (2) SATA III SSD's and presumable get 1.7x the speed instead of double (i.e. 500MB/s + 250MB/s = 1000MB/s)? OR would this result in kicking back BOTH bays to SATA II speeds? In addition, if its the 1.7x case, could you use a SATA II SSD in the optical bay and SATA III SSD in the main bay or do you have to use both the same SATA speed for raid 0 to work?

    Sorry for the lengthy question, but I tried to keep it short and simple.

    TIA!
     
  2. Orlandoech macrumors 68040

    Orlandoech

    Joined:
    Jun 2, 2011
    Location:
    Salt Lake City, UT
    #2
    You can mix/match HDDs and mx/match SSDs but doing this will be limit the RAID0 to the slowest drive and the smallest drive.

    • e.g.; Using one 256GB SATA II SSD and one 128GB SATA III SSD, you will be limited to a 256GB RAID 0 @ SATA II Speeds.
    • e.g.; Using one 256GB SATA III SSD and one 256GB SATA II SSD, you will be limited to a 512GB RAID0 @ SATA II Speeds.
    • e.g.; Using one 64GB SATA III SSD and one 64GB SATA III SSD, you will be limited to a 128GB RAID0 @ SATA III Speeds.
     
  3. GenderFi thread starter macrumors regular

    Joined:
    Sep 18, 2012
    #3
    Thanks! I think that answered my question, so in technically I will not be gaining much speed if I mix SATA II and SATA III connections. Plus I always hear reliability issues with SSD's in RAID configuration.
     
  4. Orlandoech macrumors 68040

    Orlandoech

    Joined:
    Jun 2, 2011
    Location:
    Salt Lake City, UT
    #4
    Raid0 doesn't have reliability issues because its a stripped raid, it just has no redundancy, much like a single HDD/SSD with no backup/external.
     
  5. throAU, Dec 10, 2012
    Last edited: Dec 10, 2012

    throAU macrumors 601

    throAU

    Joined:
    Feb 13, 2012
    Location:
    Perth, Western Australia
    #5
    This is not entirely true... the end result is that you're still talking about multiple components, set up in a way where failure of either one of the components will result in failure of the system.

    Your failure rate of a RAID0 volume will be approximately 2x as likely as a single disk - because you have 2 chances for an individual disk to fail.

    There's no escaping this (other than mirroring the stripe set and creating a RAID10 with 4 drives) - keep good backups. If you don't need the speed of RAID0, you'd be well advised to not use it.

    If you do need the speed (and some do), be aware of the increased probability of failure causing data loss and take the appropriate precautions.

    Just be honest with yourself, because data loss is not fun, and increasing your reliance on up to date backups in case of failure when you don't need the speed is just more work.


    edit:
    mixing/matching connections - both drives will be restricted to the speed of the slowest drive - BUT each drive only does half the work.

    So, if for example you have 2 drives:
    1 can do 100 MB/sec
    1 can do 150 MB/sec
    stripe them = the combined volume can do 200MB/sec
     
  6. GenderFi thread starter macrumors regular

    Joined:
    Sep 18, 2012
    #6
    Thanks, this clears it up even more. So it is defenitely not worth it to me as I don't need the speed and don't need twice the chance of things going wrong. I'm content with my single SSD setup. :)

    Thank you again both for clearing up my questions.
     
  7. switon, Dec 11, 2012
    Last edited: Dec 12, 2012

    switon macrumors 6502a

    Joined:
    Sep 10, 2012
    #7
    RE: RAID 0 SSDs...

    Hi all,

    Yes, I agree that RAID 0 of two SSDs roughly doubles your failure rate, but I'd like to point out that twice a really small number is still a really small number. Of course, we could argue what "really small" means, and for different people it will mean different things.

    I alway rebel against the sensationalistic press that reports on some new "DANGER!!!", such as drinking a glass of water DOUBLES!!! your risk of getting "ThatUnbelievablyLowProbabilityDisease", which will kill you. But the probability of getting this disease in the first place is 1 in 10^9, so your risk increases to 2 in 10^9 after that drink of water. At the same time, the probability of your dying in a car accident is 1 in 10^5, so you are nearly 10000 times more likely to die in a car accident than dying from "ThatUnbelievablyLowProbabilityDisease". In other words, go ahead and drink that glass of water, you aren't going to die from "ThatUnbelievablyLowProbabilityDisease" just because you drank it (even though your risk doubled --- it is still only 2 in 10^9 meaning that only 10 to 12 people in the entire world are going to die from it in your lifetime).

    (removed irrelevant info)

    Regards,
    Switon
     
  8. throAU macrumors 601

    throAU

    Joined:
    Feb 13, 2012
    Location:
    Perth, Western Australia
    #8
    Yes, it's a small number. But murphy is a prick...

    Take storage system failure rates with a bucket of salt. In theory the MTBF of hard drives should mean i've never had a failure, or that i should only have 1 in my lifetime.

    I've had multiple failed drives. Heaps of people i know who've jumped on the SSD bandwagon have had failures or firmware bugs giving them data corruption - which will potentially be enough to wipe out your RAID0.

    Just as a recent example, I just had a failure of a 2 week old enterprise SCSI drive in our new Netapp storage array.

    Failures happen. Don't believe manufacturer MTBF figures.


    That's fine for benchmarks, but are you doing anything that requires that bandwidth?

    Having used SSD equipped machines - one being an old MBA with 120meg/sec toshiba drive, and my work machine with a samsung 830, there's essentially no discernable difference in general use because things like the network, hardware detection and CPU are now the bottlenecks.

    Sure, file copies will run quicker from same disk to same disk, but if you're copying from network, usb, external drive, etc you're not going to push the RAID0 of SSD because nothing will keep up.

    But fine. Lets assume that YOU need RAID0 SSD storage throughput. Maybe you're an edge case.

    However, I maintain: if you DON'T NEED RAID0, then it is pointless increasing your risk of data loss. There's no gain.

    If you aren't an edge case and don't need more than a few hundred meg/sec IO and more than a few thousand IOPs (which is already enterprise storage array territory pre-SSD), all you're going to do is increase the probability of breaking your system and needing to go back to backup.
     
  9. switon, Dec 11, 2012
    Last edited: Dec 11, 2012

    switon macrumors 6502a

    Joined:
    Sep 10, 2012
    #9
    RE MTBFs...

    Hi throAU,

    I agree with most everything you say (see the P.S. for an exception), especially the part about MTBF numbers. Anecdotal evidence seems to suggest that MTBF numbers do not apply to actual commercially bought drives. I've always wondered about this.

    Of course, R/W errors on HDDs happen all the time, with most of these errors being recoverable. HDDs have "spare" blocks on them to handle just such errors, and these "spare" blocks are probably the reason for such long MTBF numbers. I wonder if the MTBF values are not calculated based upon the number of "spare" blocks and the average rates of recoverable R/W errors? While an actual catastrophic failure of a HDD occurs in practice not when all of the "spare" blocks are used up but rather when a specific block fails or a vibration resonance occurs that sets the flying head crashing into the rotating disk, or some such event that is not factored into the MTBF values. From anecdotal evidence, I'd have to say that MTBF numbers do not apply to actual usage in physical RAID systems.

    Maybe I am one of your "edge cases", as you so eloquently put it. But I do use the added R/W speeds and it allows me to accomplish more work faster because of what I do and how I use the RAID 0. Personally, I don't think of my usage as "edge", rather just number crunching on large data sets. I kind of resent your implications (actually rather strong) that I personally don't need the faster speeds, how would you know? If I didn't need it, why would I go to the trouble and expense of building it? The disk R/W speeds were the bottleneck for my computations before, now with a RAID 0 SSD the bottleneck is elsewhere. But I don't want to get into that here...

    And my contention still stands: double a small number is still a small number -- just because your risk for failure is double should not necessarily mean that you should not pursue the technology... If you would not pursue RAID 0 SSD because of double the risk of failure, then I would question whether you should pursue standard SSD in the first place.

    Let me give you an example illustrating my thinking about this scenario...

    Let's say that in an average number-crunching computer lifetime, say 5 years albeit this is probably on the long side, of 100 such machines having SSDs, that 1 will fail, then using a RAID 0 of 2 such SSDs on those 100 machines will increase the risk to 2 failed systems out of the 100 machines during their lifetimes. Now let's say that using the RAID 0 increases the throughput by a factor of 1.2 (2 is maximal, of course, but most codes won't run twice as fast even when the R/W speeds are twice as fast --- but put whatever reasonable number you want here, say the factor of 1.2 that I suggest - this factor depends upon exactly what code you are running). If we assume an increase throughput by a factor of 1.2 for the RAID 0 over the SSD, then during the 5 years of use we gain a full year of results (6 years of computations compressed into 5 years of time by using the RAID 0 versus the standard SSD). Personally, the guaranteed 1 year of time savings is well worth my double risk of RAID 0 failure. In other words, I would gladly trade the 2 in 100 chance of failure for the 1 year of my time, especially since it is relatively easy and no time cost for me to protect against the RAID 0 failure --- in fact, I already protect against the 1 in 100 risk of failure for the single SSD drive, so protecting against the 2 in 100 risk for the RAID 0 SSDs is not more expensive for me, neither in time nor in money. Basically, I appreciate the guaranteed gain of 1 year of my time versus the very small chance that I will have a failed drive. I think that my chances are good that I will be one of the 98 individuals that don't have their RAID 0 SSDs fail (probability is 98 out of 100), but I will have 100 out of 100 chances of gaining 1 year of computation time. One year gain in computation time can mean the difference between success and failure, the difference between publishing first and not publishing at all.

    Of course, put whatever reasonable numbers you want into the above vignette, say you believe that 5 or even 10 out of 100 machine will have failing SSDs, then the number of failing RAID 0 SSDs would increase to 10 to 20 out of 100. Still worth the risk to me because of the year I gain in computation time, and I do already protect against SSD failure with multiple backup systems. Basically, as long as the RAID 0 SSDs last long enough to complete a single computation so that I actually obtain publishable results, then I would argue that the risk of failure of RAID 0 is overridden by the gain in time.

    Regards,
    Switon

    Edit: P.S. Your comment that "murphy is a prick..." I believe does not apply here, since Murphy's Law is a compounding effect (geometric or exponential growth) while the increase in failure risk is a linear growth thing. If the RAID 0 failure was indeed exponential and not linear in the number of drives (but we both agree it is linear), then we'd have to be more careful in our analysis and your contention might just win out in the final analysis. But being linear means that my above scenario applies and not Murphy's Law. Murphy's Law is often invoked in cases where it is not applicable; where it is not a valid model for the situation at hand -- I believe that this is the case here, Murphy's Law is not applicable.
     

Share This Page