Bob - can you please translate your statements above into the "for dumbies level", does that mean the Drobo coding would not prevent using the individual sata drives in a non Drobo case or internally?? The quote from newegg.com seemed like it was the Drobo manufacturer stating that this limitation did exist. I would like to have the ability down the road to utilize the individual drives in other ways then be tied to a Drobo only device. Thanks, Don
The GetDPI Photography Forum
Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!
DROBO Storage device
- Thread starter Jack
- Start date
I think basically what Bob is saying is ANY RAID 5 array uses special coding for the drive array, so that if you pulled any single drive from that array you couldn't read any data from it either. IOW, any single drive from any RAID array, DROBO or otherwise, is for the most part not going to be readable once removed from it's original array, so DROBO is not unique here.
An added note: Writes are tough in a conventional RAID 5 too unless the drive has been properly formatted for that array from the outset, and thus why you really should also keep "hot" spares if you're running a conventional RAID 5 array. (Typical RAID 5 arrays require all drives to be of the same size and often even the same manufacturer, and sometimes the same firmware or even production lots are recommended!) Since DROBO formats and builds the arrays intelligently on the fly, you can swap in mixed manufacturer, size and even speed drives whenever needed. You can even mix SATA1 and SATA2 interface drives. This latter point is a significant advantage for DROBO from a hardware overhead point -- you don't need to buy spares until you need them -- but also probably explains why the initial load times are relatively slow; the new drive needs to be formatted and the entire data ladder needs to optimally re-organized across the new array, and that process takes some time.
However, once the data is loaded and the drives all optimally organized, DROBO should run at normal throughput for the connectivity bus and probably why Marc has not noticed any appreciable slow-down on his subsequent data reads.
I'll share some specs when I've got mine up and running. Note however I am not looking to DROBO as a performance tool, but rather an efficient-redundant data storage tool.
Cheers,
An added note: Writes are tough in a conventional RAID 5 too unless the drive has been properly formatted for that array from the outset, and thus why you really should also keep "hot" spares if you're running a conventional RAID 5 array. (Typical RAID 5 arrays require all drives to be of the same size and often even the same manufacturer, and sometimes the same firmware or even production lots are recommended!) Since DROBO formats and builds the arrays intelligently on the fly, you can swap in mixed manufacturer, size and even speed drives whenever needed. You can even mix SATA1 and SATA2 interface drives. This latter point is a significant advantage for DROBO from a hardware overhead point -- you don't need to buy spares until you need them -- but also probably explains why the initial load times are relatively slow; the new drive needs to be formatted and the entire data ladder needs to optimally re-organized across the new array, and that process takes some time.
However, once the data is loaded and the drives all optimally organized, DROBO should run at normal throughput for the connectivity bus and probably why Marc has not noticed any appreciable slow-down on his subsequent data reads.
I'll share some specs when I've got mine up and running. Note however I am not looking to DROBO as a performance tool, but rather an efficient-redundant data storage tool.
Cheers,
Jack - thanks for this explanation, I understand it better now. Still don't know if a Drobo is right for me, but I can see it being so, for many. I think it would have been interesting if they could have allowed more interchangeability in its design, but guess it is not possible. For me I will look at a single external FW or FW2 2TB drive and build up from there. This may not be what many choose but it seems cheaper way to start at least...and will function for the most part in the way I am used to....old dog and all that...thanks again..
Don, totally understand and that's what my strategy was until now. My issue is when you get upwards of 3 TB of images and have all of those different external drives daisy-chained or connected on the various interface ports, it gets a little daunting keeping them all organized 
That is exactly the case and an important point, Jack. All RAID arrays have their own coding, and drives that are a part of that array must be properly formatted for the array. This can even be true for RAID 1 (mirror) arrays, meaning that if one drive in the array fails, the other drive will rebuild the array with the replacement. It can be a false security thinking that if one failed, you still have all your data readable from the other outside of the RAID array configuration. Only way to ensure that is to clone one drive to another as pure standalone drives.
The performance on the Drobo should be pretty decent, once it is allowed to format the drives, and get itself set up properly. This will take time, but then it takes time on any RAID array, so folks should not be surprised. (The videos are a bit misleading when drives are popped in and out. That part is possible, but the full performance capabilities will not be there until the system gets completely stabilized.)
Jack, using this device for efficient-redundant data storage would be ideal, but I also think it could be used in normal storage workflow once drives are in place, formatted, and the system is stabilized. The FW800 capability is more for date throughput, and will be "gated" somewhat by the internal processing. The USB2 channel will be a lot slower, as USB goes through the CPU. In this device, they are essentially installing a small CPU for that traffic, so the hit within the device will not be bad, but USB2 back on the main computer will be a different issue, and will be affected by the CPU usage of that computer. (This is one of the nicer things about FW....it is more or less able to handle data transfers without involving the main CPU of the host, and that is why it can have higher sustained throughputs than USB2.)
So folks do need to understand that this device has some great capabilities, and that it is also not going to be as fast as hardware RAID arrays, and that drives placed into its system will take time to build and organize. If you can live with those parameters, and let it set itself up properly, it should work very nicely and provide some level of confidence for data redundancy.....to a point.
LJ
The performance on the Drobo should be pretty decent, once it is allowed to format the drives, and get itself set up properly. This will take time, but then it takes time on any RAID array, so folks should not be surprised. (The videos are a bit misleading when drives are popped in and out. That part is possible, but the full performance capabilities will not be there until the system gets completely stabilized.)
Jack, using this device for efficient-redundant data storage would be ideal, but I also think it could be used in normal storage workflow once drives are in place, formatted, and the system is stabilized. The FW800 capability is more for date throughput, and will be "gated" somewhat by the internal processing. The USB2 channel will be a lot slower, as USB goes through the CPU. In this device, they are essentially installing a small CPU for that traffic, so the hit within the device will not be bad, but USB2 back on the main computer will be a different issue, and will be affected by the CPU usage of that computer. (This is one of the nicer things about FW....it is more or less able to handle data transfers without involving the main CPU of the host, and that is why it can have higher sustained throughputs than USB2.)
So folks do need to understand that this device has some great capabilities, and that it is also not going to be as fast as hardware RAID arrays, and that drives placed into its system will take time to build and organize. If you can live with those parameters, and let it set itself up properly, it should work very nicely and provide some level of confidence for data redundancy.....to a point.
LJ
Agreed LJ: IF your RAID array looses a drive, you need to install a new drive and let the array repair itself to re-claim the data. In the event of some array hardware failing, you have two choices:
1) buy the same piece of hardware you had before and reinstall the drives in it. With DROBO this is confirmed to work if both units have the same firmware, but probably varies from yes to maybe to no with other hardware RAID systems;
2) have your data redundantly backed up on some other system.
Personally I like option 2 the best
1) buy the same piece of hardware you had before and reinstall the drives in it. With DROBO this is confirmed to work if both units have the same firmware, but probably varies from yes to maybe to no with other hardware RAID systems;
2) have your data redundantly backed up on some other system.
Personally I like option 2 the best
Jack - guessI am not there yet, so far have 2-3 500-750 gb ext drives backing up redundantly the same 400gb image files and other business files. For my purposes, having a 2TB drive will suffice for now, and can add a second or third when needed, or a RAID or Drobo when I have run out of room. By then, maybe more options will be available. For the cost of a Drobo and 2 1TB drives to start, I can buy two 2 TB ext drives, so for me the cost part is still in favor of single ext drives like I have had. With the release of USB 3.0 or FW1600/3200, I guess I will be able to add 1 without having to replace the Drobo USB with the FW model. I assume the newer Drobo will be only FW400 or 800 at most. The external drives being available even know can be FW800 compatible. Can anyone comment on the speed of such drives with FW800?
Jack,
Thanks for the translation, yes that is what I meant.
However, lets say that you have a four drive RAID 5 array. Any three of those four drives could be moved to another from of the same model and revision as the original frame and it should work in "degraded mode" (without the ability of sustaining another error), or if all four drives were moved then everything would work fine. Note that once a drive has been removed from a raid 5 array and even if a single write has occurred no matter how small to the remaining drives, then the drive that was removed must be re-built even if it is replaced in its original position. EVEN IF NO WRITES HAVE OCCURRED AT ALL, the array must be re-scanned completely to determine that all of the distributed crcs are correct since the disk control has no idea where that disk has been
. Rebuilds or re-scans take a long time, on the order of a day per terrabyte on some systems, during which time system performance will be degraded due to the busyness of the disks and reliability features will not be present until the rescan or rebuild completes.-bob
FW 800 has a theoretical throughput of 800 Mbit/s, or about 100 MByte/s. While single SATA2 drives conected via SATA2 ports have burst speeds up to 3Gbit/s using the buffer, the fastest (single) SATA 2 drive can only sustain a little over 100 MByte/s on large read/writes, and even they slow down to below 80 MByte/s as they pass to the slower half of the drive. Striping (RAID 0) can increase the throughput significantly, theoretically increasing it by a factor equal to the number of drives striped, however reliability is reduced by a factor of 1-(1/D).
On my system, I have confirmed sustained FW 800 throughput at around 80 - 90 MByte/s with previous generation SATA2 drives, or roughly at about their maximum sustainable transfer rate. I have not tested the newest drives, but suspect they will approach or slightly exceed the theoretical FW 800 maximums.
So at least for now, FW 800 is about as good as it gets for large file transfer speeds on a single SATA2 drive and probably why there is no big push for FW1600. The only way to improve throughput at present is with a RAID scheme on multiple drives.
PS: FWIW, I have confirmed virtually identical throughput speeds to those above on single drives across GigLan.
Jack - many thanks that certainly cleared things up per my needs at this point, am comfortable going with 2TB and FW800 for now. I am always appreciative about the wealth of info available here. Best of Luck with the Drobo etc. It certainly looks like the best option if need be. Don
I bit the bullet and ordered one of these. It seems to have the added advantage of being very simple to use for those of us who are less expert with regard to RAID setups. Now I have to decide which drives to order for it. Since it's primarily for backup I'm wondering if I need the latest and greatest (and most expensive) drives.
Daivd, I'm putting in a set of mis-matched 1TB drives, 2 WD's and 2 Samsungs. But I still don't have it yet!
When I ordered it, the site said "available". The next day it said estimated 10 days delivery. Yesterday it said 5 days. I ordered it over a week ago. Customer service says they cannot give me a specific delivery because they don't have a solid date yet.
So far I CANNOT say I'm at all impressed with their service. Starting to sound like your Rollei experiences...
When I ordered it, the site said "available". The next day it said estimated 10 days delivery. Yesterday it said 5 days. I ordered it over a week ago. Customer service says they cannot give me a specific delivery because they don't have a solid date yet.
So far I CANNOT say I'm at all impressed with their service. Starting to sound like your Rollei experiences...
The Seagate 7200.11 are excellent, but run a bit warmer than the WD Green or Samsung Spinpoint, a consideration when 4 are packed tightly together in a small box...
One of the reasons I bought the DROBO was I already had a bunch of mix and match drives. Best thing about the DROBO is you can add drives as you need them, so no sense buying more than you need for the present -- one certainty to keep in mind is that the price of drives will continue to drop and the technology inside them will continue to improve...
One of the reasons I bought the DROBO was I already had a bunch of mix and match drives. Best thing about the DROBO is you can add drives as you need them, so no sense buying more than you need for the present -- one certainty to keep in mind is that the price of drives will continue to drop and the technology inside them will continue to improve...
D
ddk
Guest
Drobo is fine if you have a bunch of mismatched drives laying around collecting dust but its not what you buy if starting from scratch and going for the latest/greatest HDs. Drobos are going to be slow no matter what the interface, its inherent to their design and its expensive doing it your way. You're much better off with a high quality, high speed e-sata RAID from LaCie or other competing manufacturers. Drobo's bottleneck is copying large files over, which is going to be the case if you're going to use it as your main image server, specially in your case David and the type of files you're dealing with. Setting up a RAID array is very simple, almost as simple as formatting a disk. You open he software, decide on the RAID type, there's a clear and detailed explanation of each type with the software, then click format, that's it, the rest is invisible.
Jan Brittenson
Senior Subscriber Member
Likely there's no designated parity spindle, and instead they use fountain codes to achieve redundancy.
http://en.wikipedia.org/wiki/Fountain_code
Jan Brittenson
Senior Subscriber Member
Do they allow setting the degree of redundancy?