amsung Pro 500 2x 512GB or 1x 1TB

[davidst99]

Hi, I'm looking at geting a Samsung Pro 500 SSD drive. Would it be a good idea two get 2x 512gb running in RAID 0 or buying 1x Samsung Pro 512gb. Thanks.

David

 

[hhhd1]

RAID 0 is pointless for SSDs

 

[Coup27]

Only in highly sequential workloads such as professional audio/video editing or photo editing will you receive any benefit from SSD's in RAID0.

Unless you are running that type of workload get a single 1TB.

 

[Diogenes2]

RAID 0 is pointless for SSDs

Depends on what your point is...

 

[CiPHER]

RAID 0 is pointless for SSDs

Ahh, i cannot stand this! Please do not take is personally, but...

SSDs owe their speed thanks to RAID0! Modern SSDs are like 16 devices in RAID0, or RAID5 more often, but that is beside the point.

What may be pointless may be an even faster storage device! But RAID0 does pretty well in increasing performance.

If the SSD does it internally, you can amplify the effect externally with host-level RAID0. Works just great to boost the I/O scores. Won't do much in practical experience because a single SSD performance is already way high to cap the CPU most of the time. Storage no longer is a huge bottleneck, except for special tasks. But common usage a single SSD is just fast enough. You don't need more. You don't run a huge database where you need a million I/O operations per second. A desktop is just that; a system designed for a single user, not a server that services thousands of clients over the internet. That is where you might need more than just one SSD in terms of performance.

Because even how great RAID0 is, it has one weakness: it cannot improve random read IOps with a single queue depth. And this is the most important performance spec for a desktop user. Booting, launching applications, many things will rely on this performance spec. And virtually all SSDs do between 20 and 30MB/s. This is because it cannot benefit from the principle behind RAID0 - interleaving. Because consequent I/O requests cannot be predicted like it can with sequential I/O. The result is that it is fully latency-bottlenecked. The NAND latency, the SATA latency (propagation delay), the OS overhead. Your CPU speed will affect the results, the controller will, and finally the SSD will too. But in the end, all SSDs do between 20 and 30MB/s of random blocking 4K read performance. And this is the most important performance spec for desktops! Funny that all consumer look at the high numbers; they are not nearly as interesting as the lowest number - ha ha! Typically how i encounter these things throughout the industry...

Now back to the subject: RAID0 is great but ask yourself: why do you need an even faster SSD? If just for kicks, then sure. Maybe you could split them sometime and use them for 2 desktops for family/whatever? But don't think your games will run twice as fast, or your boot performance rocks, or something like that. It will not exceed the threshold needed to exceed Placebo-experience. Which is about 25-40% of actual improvement. It will be more like 2 - 18%.

 

[hhhd1]

Depends on what your point is...

IMO, if you need to ask, you probably do not need it.

You get an increase of sequential transfers, with decreased reliability, and depending on the controller, you may be loosing TRIM, which could result in further performance decrease.

Enabling Samsung's RAPID mode is better overall solution.

 

[CiPHER]

RAPID is a pile of junk intended to play on consumers getting horny on high numbers... It has no real relevance in practice. You have an even faster write-back VFS cache built-in your operating system. The scores just don't show that.

Basically, RAPID shows you RAM scores, not SSD scores. It also rapes you reliability with an insecure write-back which increases susceptibility to corruption. I high advise NEVER TO USE RAPID!

 

[hhhd1]

Ahh, i cannot stand this! Please do not take is personally, but...

SSDs owe their speed thanks to RAID0! Modern SSDs are like 16 devices in RAID0, or RAID5 more often, but that is beside the point.

I do not disagree with you, but this is a general question, and I was assuming that the question is regarding typical work loads, (office application + web + games + multitasking + page file)
My assumption could be wrong, but if not, then I stand by my answer.

Now back to the subject: RAID0 is great but ask yourself: why do you need an even faster SSD? If just for kicks, then sure. Maybe you could split them sometime and use them for 2 desktops for family/whatever?

That is a very good point, I didn't think of that.

 

[hhhd1]

RAPID is a pile of junk intended to play on consumers getting horny on high numbers... It has no real relevance in practice.

And that is the same reason why many run SSDs in RAID 0, just to see sequential benches that are twice as fast, with no real benefit in their workloads.

 

[CiPHER]

And that is the same reason why many run SSDs in RAID 0, just to see sequential benches that are twice as fast, with no real benefit in their workloads.

The same applies to a single SSD. If not for RAID0 it would be reduced to USB-stick speed ranges. Those often have only one or two channels.

Furthermore, you seem to suggest RAID0 only increases sequential throughput. That is not true; it increases random IOps too. That is why the QD32-score is more than 10 times as high; that is all thanks to the principle behind RAID0.

It is funny that people want faster SSDs and look a lot of benchmarks, but when they hear RAID0 they don't want it. I just found it funny that the SSDs they so adore achieve their high speed thanks to RAID0 - or interleaving whatever you wish to call it.

The reality is that a single SSD (a RAID0 of 16 NAND) is already fast enough for desktop systems. An even faster SSD will not provide real improvement because the CPU is bottlenecked, not the storage. So to improve performance you would need a 10GHz processor or something of sorts. Then you make the CPU-bottleneck smaller allowing more SSD utilization instead, which in turn would make a faster SSD worthwhile again. But as it stands now, a single SSD is already very fast and more than enough for a desktop system.

RAID0 is 'free' though; the price of 2x512GB is about the same as 2x1TB. Do not RAID Samsung SSDs though; get a proper Crucial MX100 to do that. The 512GB version also has almost capped write speeds, so it would be cheaper and better in multiple ways.

Good luck!

 

[Coup27]

_CiPHER_ you are clearly very knowledgeable but I do think you have lost perspective in this discussion. You have spent the majority of the time discussing the internal RAID construction of an SSD or the inner workings of things which aren't actually relevant to the question being asked.

Putting two SSD's in RAID0 will only significantly improve 2 metrics which are sequential read and sequential write and only specific workloads such as what I indicated in post #3 will take advantage of that. For the vast majority of users who don't run those workloads putting two SSD's in RAID0 yields very little you can actually appreciate without a benchmarking program which is why people say it's a waste of time.

I ran 2 Samsung SSD's in RAID0 for 6 months without any issues at all so I would like to hear why you say that you should not run Samsung SSD's in RAID and get a "proper" Crucial drive.

 

[CiPHER]

_CiPHER_ you are clearly very knowledgeable but I do think you have lost perspective in this discussion. You have spent the majority of the time discussing the internal RAID construction of an SSD or the inner workings of things which aren't actually relevant to the question being asked.

You may have a point here.

But i did gave some practical advice. Having 2x 512GB SSD in RAID0 would provide little practical performance benefit, but it is around the same price and any performance increase is welcome. The 1GB/s read might speed up loading games a bit, or other tasks. And you can break them up into 2x512GB for two other systems for family/girlfriend/bedroom etc.

Putting two SSD's in RAID0 will only significantly improve 2 metrics which are sequential read and sequential write

That is simply incorrect. It improves both random read IOps with multiple queue depth, and random write IOps - even with single queue depth.

Let's look at CrystalDiskMark on a single SSD :

Both the sequential read and write are so high, thanks to RAID0. Actually many SSDs internally could do about 2GB/s of sequential read, if internally there were enough bandwidth/channels available and the SATA interface is traded for PCI-express. Some OCZ PCIe products have benchmark results where this can be seen - yes with incompressible data.


Random read
"4K" means random I/O with a request size of 4 kilobytes. But it also means that the queue depth will be just 1. In normal English: it only sends the next I/O request after the SSD has returned the data of the previous read request. In other words: it can only do one thing at a time, and not do multiple things at a time like is possible with RAID0. And this is exactly why the SSD cannot employ its internal RAID0 to boost the performance. That is why 4K random read is always between 20MB/s and 30MB/s - dependent on CPU speed as well. This is fully latency bottlenecked. Round trip times. Like gaming, you need a low ping; i.e. fast reaction time; i.e. low round trip time.

"4K QD32" means random I/O of 4KiB request size, with a queue depth of 32. That means 32 blocks of 4KiB will be asked from the SSD at the same time. SATA allows a maximum of 32 queued I/O's, only SAS can go higher. This means the SSD has a maximum chance of doing 32 times at once, if internally it is able to do that. Otherwise, the SSD just let the request wait and let it complete later with higher latency.

The higher the queue depth for random reads, the more the SSD can enable the power of the internal RAID0 or interleaving design where one NAND die processes another I/O than the other NAND die. Basically having a 16-disk RAID0 internally.

With all inefficiencies combined, you can see the score is only 10,6 times higher, and not 15 or 16 (RAID5/RAID0). Still, this is the power of RAID0. It is an awesome piece of technology that works extremely well.


Random write
Unlike random read, random write can be accelerated by RAID0 even with a queue depth of 1 - meaning blocking writes. With reads this is not possible, because the SSD has to finish the read request first. But with writes, you can make them disappear with write-back buffering. You store the request in your command buffer, you tell the host that you've completed the request, and internally you can queue the write request.

This all means that even with the host sending one random write at a time, the SSD can process multiple at once. This works because the SSD immediately returns I/O completion and only processes the write later with multiple writes at a time.

Why else do you think the random 4K write is so much higher than 4K read? SSDs do writing a lot slower than reading. But thanks to buffering the write, the SSD can execute multiple 4K blocks at once, thanks to RAID0.

Theoretically, a single queue depth for writes could be enough to saturate all performance potential. It is for HDDs, which uses the exact same principle discussed above by the way. HDDs cannot process multiple writes at the same time, so their 4K write will be about the same as 4K-qd32. But for SSDs, a single queue depth for random writes is not enough. SSDs being so fast, they get starved for I/O requests because each I/O request has a propagation delay. So you can only send x amount of I/O requests per second. So you need multiple random write queue depth to saturate performance. Between 4 and 8 probably would be very close to 32 already. This doesn't count for random reads, they cannot 'hide' read requests; you can only do that trick for writes.

The screenshot above shows only a x2 improvement between queue depth 1 and queue depth 32, which is exactly in line with the theory i explained above.


Final verdict
So there you have it. A single SSD uses the principle behind RAID0 very effectively to boost all but one of its performance specs. Only the "4K" random blocking read would not allow itself to be boosted by RAID0.

The funny thing is that i explained RAID0 performance characteristics by showing a benchmark of a single SSD. But the very same performance behaviour applies to host-level RAID0 like Intel onboard RAID. Because it is the same principle. That is why i think it is so funny people love fast SSDs but dislike RAID0. Haha. :awe:

I ran 2 Samsung SSD's in RAID0 for 6 months without any issues at all so I would like to hear why you say that you should not run Samsung SSD's in RAID and get a "proper" Crucial drive.

Samsung SSDs employ journalling on the mapping tables. This allows them to omit the capacitors that Crucial employs to protect its FTL-consistency. In normal English: both have the same goal, but use different means. The Samsung uses a 'cheap' software protection, cheap in the sense that it costs no extra parts such as Crucial's SSD. It's simply a design-choice for the rather complex firmware that basically is the operating system of the SSD.

The SSD is a mini-computer with up to three processor cores, its own RAM and its own operating system (firmware) and well you know all that. But my point is its complexity. SSDs are not simple storage devices like harddrives. They need some form of protection or the NAND flash is unprotected, and this is what caused 90% of OCZ SSD failures in the past; the firmware gets stuck after the NAND is in an inconsistent state where the mapping tables do not match the actual stored data. A secure erase would 'fix' the issue at the cost of total data loss. Many consumers sent their SSD to OCZ and got a secured erased/wiped version back, many times from a different customer having the same issue. Better protection against power loss and keeping the mapping tables consistent, was obviously the next step.

The software protection that Samsung SSDs employ, can cause a form of inconsistency that RAID and modern filesystems like ZFS were never designed for. This has to do with the protection 'rolling back' the mapping tables to an earlier state. In essence, the SSD is reset to a point in the past, with all LBA corresponding exactly like it was in the past. But not all SSDs may roll back after a power failure, or roll back to different times in the past. Then you have an inconsistent filesystem.

After a power failure, a Samsung SSD will employ POR (Power-On Recovery). You can see how many times in the SMART attributes. This protection would roll back the mapping tables to an earlier state. This means any recent writes which may be corrupt on the NAND after a power failure, will not be used at all. Clever trick for sure.

But it has a weakness. Imagine a RAID array where disk1 is of another date than disk2. If all SSDs roll back to the exact same time, there would not be a problem. But if parts of the filesystem that spans multiple SSDs are set back to the past, you would have inconsistency in your filesystem.

The protection works flawlessly on simple desktops though, because in that case the whole filesystem is put back to the same date. It is cheap and effective. But on more complex storage, i would recommend avoiding Samsung SSDs. With a legacy filesystem like NTFS, you would not notice there is corruption, you just see the filesystem check for a very brief amount and assuming Windows still works, you might have some missing files or some corruption in your game that you were updating. It doesn't have to be serious though. But these kind of problems could give you weird issues where you might not expect the SSD being the culprit. With the Crucial MX100 being even cheaper, the choice is pretty obvious if buying a new SSD for a RAID or ZFS setup - in my opinion of course.

 

[BonzaiDuck]

Insightful post. You have enlarged . . . my mind! ..

In the human perception, your eye and brain cells don't work fast enough to register the difference between -- say, sequential -- throughput of 490 MB/sec and throughput of ~900 MB/sec or more. Making that bottleneck even wider than it already is should be quite pointless.

But if you want to get an equivalent speed-increase for certain pervasive user profiles, you could use SSD and even RAM-caching to speed things up, even though YMMV depending on the file attributes, size and disk-management factors.

The revelation is that you don't have to do it by choosing a limited-application "proprietary" solution like ISRT, Marvell Hyper-Duo, or Samsung RAPID. Nor do you need be limited to a particular storage mode, like ISRT (or RAPID -- possibly Marvell. I can execute the caching for multiple hard disks, either directly to RAM or to a caching 60GB SSD, while I can also cache one or more SSDs to RAM.

Other than Primo-Cache, the only other similar software I know of would be Super-Cache -- former from Romex Software, latter from Super-Speed LLC in Massachusetts. Romex HQ is in Shanghai.

Other than the People's Army wanting to read my mail or check me out by hacking my web-cam and paranoid fears that Romex has some sort of Yellow-Peril worm or bug, you get the most-est for the least-est -- a sort of Swiss-Army knife of caching across three levels in the hardware pyramid of storage.

Ahh, i cannot stand this! Please do not take is personally, but...

SSDs owe their speed thanks to RAID0! Modern SSDs are like 16 devices in RAID0, or RAID5 more often, but that is beside the point.

What may be pointless may be an even faster storage device! But RAID0 does pretty well in increasing performance.

If the SSD does it internally, you can amplify the effect externally with host-level RAID0. Works just great to boost the I/O scores. Won't do much in practical experience because a single SSD performance is already way high to cap the CPU most of the time. Storage no longer is a huge bottleneck, except for special tasks. But common usage a single SSD is just fast enough. You don't need more. You don't run a huge database where you need a million I/O operations per second. A desktop is just that; a system designed for a single user, not a server that services thousands of clients over the internet. That is where you might need more than just one SSD in terms of performance.

Because even how great RAID0 is, it has one weakness: it cannot improve random read IOps with a single queue depth. And this is the most important performance spec for a desktop user. Booting, launching applications, many things will rely on this performance spec. And virtually all SSDs do between 20 and 30MB/s. This is because it cannot benefit from the principle behind RAID0 - interleaving. Because consequent I/O requests cannot be predicted like it can with sequential I/O. The result is that it is fully latency-bottlenecked. The NAND latency, the SATA latency (propagation delay), the OS overhead. Your CPU speed will affect the results, the controller will, and finally the SSD will too. But in the end, all SSDs do between 20 and 30MB/s of random blocking 4K read performance. And this is the most important performance spec for desktops! Funny that all consumer look at the high numbers; they are not nearly as interesting as the lowest number - ha ha! Typically how i encounter these things throughout the industry...

Now back to the subject: RAID0 is great but ask yourself: why do you need an even faster SSD? If just for kicks, then sure. Maybe you could split them sometime and use them for 2 desktops for family/whatever? But don't think your games will run twice as fast, or your boot performance rocks, or something like that. It will not exceed the threshold needed to exceed Placebo-experience. Which is about 25-40% of actual improvement. It will be more like 2 - 18%.

 

[aviator79]

Guys, there was a question in the beginning of this thread. Your discussion is pointless.
You tell some good features of what you thing is best, but you forget to tell the negative thing.
Back to topic: Nobody can tell if it is a good idea for YOU, because you did not say, what you want to to with your SSDs.
Raid
PRO: faster speeds...but not noticable in every kind of workload
CON: If one SSD fails, your datais lost, slower access times
Single:
PRO: Reliable, easier to manage
CON: Slower...but not slow
I personally like to use some medium sized SSDs, but not in RAID. So I am more flexible, can replace one, one can fail, I still have some other, I can switch them to other PCs/Laptops...

 

[Diogenes2]

CON: If one SSD fails, your datais lost,

If one SSD/HDD fails in any configuration your data is lost..

 

[davidst99]

If one SSD/HDD fails in any configuration your data is lost..

I understand there's a risk of failure is greater in RAID 0. I do an incremental backup every night on two WD 4TB Black drives in RAID 1.

 

[CiPHER]

If one SSD/HDD fails in any configuration your data is lost..

If one NAND chip fails, the whole SSD is dead. Does this mean you dislike SSDs in general? Probably a multi-core CPU will have a higher chance of failure than a single-core CPU. But i never heard anyone say: "ohhh don't do that, its less reliable!"

The doubling of failure rate with RAID0 is over-estimated. Whether it fails twice as often doesn't change the fact you need to backup stuff you wish to keep. And when backed up, the risk is degraded to only a hassle - lost time and energy spent on reinstalling and restoring the backup. You don't want that to happen too often. But whether you run RAID0 or not, if you store things on your SSD you wish to keep, you ought to make backups. Fortunately, SSDs are easily backed up due to their limited capacity. Fortunately, in many cases the data on an SSD is the data you do not wish to keep. Installed files, installed games. You do want your profile directory with all saved games etc. backed up. But a lot of stuff on your average desktop SSD doesn't need a backup.

Again, these myths of the past about RAID0 are preventing a fresh objective look on the subject. RAID0 is not evil, it is useful. Its disadvantages are clear and not all that bad, its advantages are not very clear, but actually are rather good. Funny huh.

 

[Byte]

Not only can a single drive fail, there could be RAID 0 Failures in the drive, but i've also encountered many RAID failures in software/driver, OS, or controllers. In fact probably more often than actual drive failures. If you're not up with dealing with that, i'd stick to single drive.

 

[hhhd1]

If one NAND chip fails, the whole SSD is dead. Does this mean you dislike SSDs in general? Probably a multi-core CPU will have a higher chance of failure than a single-core CPU. But i never heard anyone say: "ohhh don't do that, its less reliable!"

You are comparing apple to oranges, when one block fails in an SSD, which happens too often, the SSD try to recover the data using ECC and/or RAIN, and then the Block is replaced with another block from the reserved space.

In some cases where the error is too big to recover via ECC or RAIN, an IO error is sent to the OS/application.

With SSD (or processors), single chips (or cores) are usually reliable, there hasn't been that much reports of complete data loss or failure because of them, unlike with RAID0, where allot other software and hardware factors could result in breaking the RAID0 array.

Based on what you said about Samsung's software based protection against power failure, it seems that doing RAID0 using Samsung SSD is specially a bad idea.

IMO, if you really need RAID0 for your specific workload, make it RAID10. and before that, consider RAMDISK.