Will PCI-Express-Based Solid-State Drives Become More Popular in the Near Future?

Solid-state drives are an amazing innovation in the realm of computing, because of their amazingly fast transfer speeds and very low power consumption in comparison to mechanical hard drives. They eliminate the slowest component of a computer, making its overall operation far faster and more efficient.

Currently, most solid-state drives use the SATA interface, as that is the most commonly-used interface for traditional hard drives. That is understandable, as that allows the drives to take advantage of existing infrastructure, without needing to introduce an entirely new interface for them. However, it is almost certain that SSD's shall advance to the point that the SATA interface is not sufficiently fast for them, requiring that they use an even faster interface.

Such an interface would be the PCI-express interface, in my mind. It has far greater data transfer rates than does SATA, and also can power a device itself, eliminating the need for additional cables. At the present time, there are solid-state drives that use this interface, but they are not commonly used and currently are more expensive than their SATA-based counterparts, due to the cost of the interface. However, I am certain that such devices shall become less expensive, and thus, used more frequently, as time passes and they can be manufactured more efficiently.

What does everyone else say on this subject? Shall PCI-express-based solid-state drives gain greater usage in the near future? I await your responses.

I see it as a benefit since the PCIe bus has far greater bandwidth. The only drawback is adoption. Sure, it may be the greater of the two options, but if it's not adopted in the industry it will never meet the manufacturing quotas to bring it into the price point for wide adoption. It's like the VHS/Beta wars again.

For now, I think SSDs aren't fast enough to even saturate sata 6Gb/s, so there's no point with PCI. Also, I think most people who are spending lots of money on SSDs are also likely to be getting nice video cards, which take up lots of slots on the board.

Idk though, I could be wrong.

My company sells a lot of the Texas Memory Systems PCIe based SSD modules (RamSan). The main drawback is the expense.

We also sell their standalone systems and they can get into the 6 digit price range...

Once these systems become more mainstream, the price will come down. IBM sees this and that is why they recently bought TMS. I think it is going to be at least 5 years before the price comes down enough to make them a real consumer option.

I'm guessing that the future holds faster SATA speeds, making PCI-E SSD's more or less irrelevant. They're just too large and typically take up wanted space inside of PC's. There will probably always be a market for them, though.

Quote:

---

Originally Posted by CrazyTeaPot

I'm guessing that the future holds faster SATA speeds, making PCI-E SSD's more or less irrelevant. They're just too large and typically take up wanted space inside of PC's. There will probably always be a market for them, though.

---

First, while I am certain that a faster version of SATA may eventually be developed, PCI-express shall always be the faster interface, because of its nature, which allows for it to scale far better than can SATA.

Second, the PCI form factor is actually better than the SATA form factor in that it provides greater uniformity to components of a computer: since video cards, sound cards, and other add-on cards already use that form factor, it is only logical to manufacture data storage devices that utilize that form factor, as well, since SATA-based devices do not typically occupy the same physical space as do the central components of a computer (processor, motherboard, memory, and add-on cards).

Therefore, I definitely believe that PCI-express-based solid-state drives shall definitely increase in popularity and feasibility in the future; the only uncertainty is when and how that shall occur.

Quote:

---

Originally Posted by DemonDragonJ

PCI form factor is actually better than the SATA form factor in that it provides greater uniformity to components of a computer: since video cards, sound cards, and other add-on cards already use that form factor, it is only logical to manufacture data storage devices that utilize that form factor, as well, since SATA-based devices do not typically occupy the same physical space as do the central components of a computer (processor, motherboard, memory, and add-on cards).

---

While I don't disagree, the problem is that I can get a few video cards in my machine, and then I don't have room for anything else.

I really think the days of dedicated sound cards, ram drives, and PCI based SSDs are limited by the nature and size of our motherboards. If I have a choice between PCI based SSD (which is still prohibitively expensive) and another video card, I'll choose having another video card and a sata SSD.

Part of the cost of PCI SSDs are the fact that they use a different kind of memory that can actually use the speed of the PCI slot.

Standard SSDs are getting cheaper because of scale, but it's also cheaper memory. There are very few drives that can even saturate a 6gb/s sata port as it is. The drives that can are very expensive, and the same will hold true for PCI based drives.

I would love to have one as I am still using an X58 system and it only has 3GB sata ports. I would have to loose my Physix card but I think it would be a lot faster than connecting the card through sata. My only question is how will it effect the video bandwidth on the PCIE lanes?

I think it's obvious what's going to happen just by going by current popularity and information.

The price difference is fairly minimal, yet SATA is still winning the race by miles. With SATA, you can get 240GB's at 465MB/s Write for $210. With PCI-E, you can get 240GB's at 900MB/s for $280. The performance increase is quite large, but people are still mostly buying SATA.

The variety alone when comparing the two is another large variable in this situation. OCZ is pretty much the only company currently offering PCI-E SSD's, yet despite that they have the entire niche to themselves, they aren't really selling that well, at least not in comparison to SATA drives. So why invest? It's not a, "If you build it, they will come" situation, because it's built and still people are only minimally interested. (Like I said, it will always exist, just in smaller numbers.)

PCI-E SSD's take up room in PC's in multiple ways too, which most people aren't going to want. You basically have to have the right motherboard layout in order for it work. But then, on top of that, you have to also consider that there's a very real chance that it will disrupt airflow in that general vicinity, which will affect GPU's and CPU's alike. It's a bit of a problem unless you build with it in mind specifically, which might not be possible or desirable for a lot of people/companies.

SATA is universal, simple and already widely established. Why would they change? Because PCI-E is twice as fast as current SATA models? That's something to consider, but most people seem to be happy with their already very fast SATA drives. Even the old Vertex 2 Drives are pretty fast, at half the speed of the Vertex 4 Drives. The current benefit of going PCI-E is just negligible to your average consumer. The only difference that they're going to notice is an increase in cost and possibly frustration due to the things mentioned above, so why do it?

It's an enthusiast market, just like those WD Raptor drives of the past. They'll exist, without a doubt. But it's not going to take off like the other drives do. It's a niche market product, not a mass-market one.

Quote:

---

Originally Posted by Mechh69

My only question is how will it effect the video bandwidth on the PCIE lanes?

---

I have often heard computer users speak of "lanes" when discussing PCI and PCI-express, but, to the best of my knowledge, PCI-express is a point-to-point interface, rather than a multi-drop interface, as is conventional PCI. I am not an expert, but should that not mean that each port, and thus, each device in each port, has its own dedicated connection the CPU, which means that they do not share lanes, and thus, a device in one port shall not detract bandwidth from another? Is that the case, or am I misunderstanding how PCI-express functions?

Quote:

---

Originally Posted by DemonDragonJ

I have often heard computer users speak of "lanes" when discussing PCI and PCI-express, but, to the best of my knowledge, PCI-express is a point-to-point interface, rather than a multi-drop interface, as is conventional PCI. I am not an expert, but should that not mean that each port, and thus, each device in each port, has its own dedicated connection the CPU. which means that they do not share lanes, and thus, a device in one port shall not detract bandwidth from another? Is that the case, or am I misunderstanding how PCI-express functions?

---

Take a look at the traces on the mobo.

Quote:

---

Originally Posted by DemonDragonJ

I have often heard computer users speak of "lanes" when discussing PCI and PCI-express, but, to the best of my knowledge, PCI-express is a point-to-point interface, rather than a multi-drop interface, as is conventional PCI. I am not an expert, but should that not mean that each port, and thus, each device in each port, has its own dedicated connection the CPU. which means that they do not share lanes, and thus, a device in one port shall not detract bandwidth from another? Is that the case, or am I misunderstanding how PCI-express functions?

---

In his case, the x58 northbridge chip is responsible for 32 pci-e lanes (which are going to be routed to the graphics card slots) and the 6 lanes being handled by the southbridge. The southbridge connects to the northbridge by DMI bus and the northbridge connects to the processor by quick path bus. The quick path connection allows for 25.6 gigabytes per second. The 32 lanes of pci-e on the northbridge can use a max of 18 gigabytes per second. The DMI connection to the southbridge is 2 gigabytes per second.

The graphics slots should get their full bandwidth. The slots on the southbridge total 3 gigabytes per second and the connection to the northbridge is only 2 gigabytes per second, so they could technically fight for bandwidth.

I have noticed that PCI-express-based solid-state drives often have their flash memory cells in a RAID 0 configuration, and I wonder why that is. I am not an expert on RAID, but I do know that, while RAID 0 does provide an increase in data transfer rates over a non-RAID configuration, it also has no redundancy and thus is very vulnerable to irreversible data loss, and a flash memory device with a PCI-e interface is already so fast that I do not believe that it would need to be in a RAID configuration. What insight can anyone else here offer on this subject?

Cheaper to use slower cheaper flash memory in raid costs more to make faster flash non raid.

Quote:

---

Originally Posted by DemonDragonJ

I have noticed that PCI-express-based solid-state drives often have their flash memory cells in a RAID 0 configuration, and I wonder why that is. I am not an expert on RAID, but I do know that, while RAID 0 does provide an increase in data transfer rates over a non-RAID configuration, it also has no redundancy and thus is very vulnerable to irreversible data loss, and a flash memory device with a PCI-e interface is already so fast that I do not believe that it would need to be in a RAID configuration. What insight can anyone else here offer on this subject?

---

I am not sure about all of the SSD and PCIe devices out there, but almost all of the high end PCIe cards and SSD's have extra cells and in some cases entire duplication of memory chips for backup purposes. I am not sure of all the technology, but a cell failure will result in no loss of data. When I was at an Intel training class they briefly touched on the technology in the SSD's they sell and this subject came up. They were not too technical on it, but knew the basics. The same thing came up in our Texas Memory class, but it was mostly for the sales guys, so They couldn't answer (or wouldn't) the technical questions.
I do know that the Intel technology is considered a hybrid raid because of the power saving type of feature where the "spare" cell is not in use until a failure, or anticipated failure, so while it is technically part of the raid array, it may never be used. I will see if I can dig up the white paper for it.

Quote:

---

Originally Posted by Stonerboy779

Cheaper to use slower cheaper flash memory in raid costs more to make faster flash non raid.

---

Quote:

---

Originally Posted by Airbozo

I am not sure about all of the SSD and PCIe devices out there, but almost all of the high end PCIe cards and SSD's have extra cells and in some cases entire duplication of memory chips for backup purposes. I am not sure of all the technology, but a cell failure will result in no loss of data. When I was at an Intel training class they briefly touched on the technology in the SSD's they sell and this subject came up. They were not too technical on it, but knew the basics. The same thing came up in our Texas Memory class, but it was mostly for the sales guys, so They couldn't answer (or wouldn't) the technical questions.
I do know that the Intel technology is considered a hybrid raid because of the power saving type of feature where the "spare" cell is not in use until a failure, or anticipated failure, so while it is technically part of the raid array, it may never be used. I will see if I can dig up the white paper for it.

---

Both of those explanations make sense, even if they are completely different from each other. Perhaps as solid-state drives become less expensive, they shall use memory cells arranged in a traditional form?

Also, I have another question: why do most PCI-express-based solid-state drives have storage capacities in multiples of 240 gigabytes (240 GB, 480 GB, 960 GB), rather than the binary progression (8, 16, 32, 64, 128, 256, and 512 GB) that SATA-based SSD's and USB flash drives use? Those storage capacities are nearly universal, so it seems that the PCI-express-based SSD's are deviating from the normal trend with their capacities.

As for the main subject of this thread, I recently heard news about both ASUS and OCZ manufacturing new PCI-express-based solid-state drives, so I shall presume that such devices may indeed gain greater popularity in the near future.

Remember that 1GB isn't really 1,000MB... it's 1,024MB. And a 240GB drive isn't 240,000MB...

Quote:

---

Originally Posted by DemonDragonJ

Both of those explanations make sense, even if they are completely different from each other. Perhaps as solid-state drives become less expensive, they shall use memory cells arranged in a traditional form?

Also, I have another question: why do most PCI-express-based solid-state drives have storage capacities in multiples of 240 gigabytes (240 GB, 480 GB, 960 GB), rather than the binary progression (8, 16, 32, 64, 128, 256, and 512 GB) that SATA-based SSD's and USB flash drives use? Those storage capacities are nearly universal, so it seems that the PCI-express-based SSD's are deviating from the normal trend with their capacities.

As for the main subject of this thread, I recently heard news about both ASUS and OCZ manufacturing new PCI-express-based solid-state drives, so I shall presume that such devices may indeed gain greater popularity in the near future.

---

I believe that probably reserving flash cells for wear leveling that are not visible to the system. The 240GB drive probably reserves 16 gigs and then maps cells marked bad in the primary storage to the reserve chips. Likewise the 480 reserves 32 and the 960 reserves 64.

Quote:

---

Originally Posted by f8l_0e

I believe that probably reserving flash cells for wear leveling that are not visible to the system. The 240GB drive probably reserves 16 gigs and then maps cells marked bad in the primary storage to the reserve chips. Likewise the 480 reserves 32 and the 960 reserves 64.

---

This here.

To revive this thread after a long time of inactivity, ASUS has now released their RAIDR solid-state drive, which uses the PCI-express interface, and while they certainly hyped and promoted it greatly, the reviews of it that I have read have been mixed. One review stated that the device did not reach its advertised transfer rates in every test, and that a major reason for that is that the controller of the drive does not natively support the PCI-express format, so it instead must use an older SATA controller to handle data transfers.

Therefore, it seems to me that that greatest barrier to widespread adoption of PCI-express-based solid-state drives is a lack of controllers for them, so, if manufacturers begin to develop SSD controlelrs that can properly utilize the PCI-express format, will such drives become more prevalent? I myself am hoping that Samsung develops a PCI-express-based SSD, since their current SATA-based SSD's are so very awesome, and they develop every part of the drive themselves, meaning that they can test them for the best functionality. What does everyone else say about that?

Samsung already have PCIe based SSDs and have been mass producing them for mainly the ultra book and laptop market for a while now.

They do however have a new PCIe based SSD that they debuted recently but it is very much so an enterprise product. You can get it in a 1.6TB capacity with 3000MB/s sequential read speeds.

http://www.xbitlabs.com/news/storage/display/20130717231259_Samsung_Develops_Enterprise_Class_S SD_with_3000MB_s_Read_Speed_NVMe_Interface.html

I think PCI-express based solid state drives become more popular. It is a new way to adding the speed state drive to server and storage devices.

huge lol,

PCIe SSDs (in which I include M.2 SSDs and NVMe SSDs) have been available to consumers for years now. There's many offerings, but Samsung and Intel dominate the consumer market - partly because their M.2/NVMe SSDs are standard issue in many pre-built laptop/gaming/HEDT systems, partly because so many enthusiasts/gamers are willing to pay any price for top performance and brand swag. Check your local NCIX, newegg, or ebay sites.

I've been using a 1600GB PCIe2X8 CoreRise Comay E28 BladeDrive at home since early 2014. And I was a "late adopter" who actively resisted using SSDs for years.

I'm sure there's plenty of bigger, faster, better (and costlier) Enterprise PCIe SSD tech out there I know nothing about.

Holy cow, just 4 years ago, I also thought 3Gb/s was fast enough.

Now, I'm looking at investing in a RAID 0 Intel 750 U.2 setup.

To be fair, 4 years ago, M.2, and even U.2, didn't even exist. When we heard "PCIe SSDs", it was an SSD that took up a physical x1, x4, xWhatever slot, not a thin board that happened to use PCIe channeling.

Multiple PCIe SSDs in RAID0, lol.

I see new Seagate Nytro XP7200 (3.8TB and 7.7TB capacities) which use PCIe3x16 NVMe, rated up to 10GB/s sequential Read. An enterprise product at enterprise prices.

I understand PCIe Gen3 x4 is some sort of "standard volume" unit in scaling server platforms. And it's the preconfigured lane limit on M.2/U.2 interfaces. And it's a whole lot of bidirectional bandwidth, although real-world storage devices rarely approach the theoretical limit of 3938MB/s.

What I don't understand is why PCIe SSDs aren't really available in anything other than x1, x2, and x4 capacities. Many consumer motherboards have one or two x4 slots that are good for little other than a PC audio card and/or SSD drive card (and/or some kind of adapter card, if your motherboard somehow doesn't already have enough dozens of fast USB ports and gigabit network plugs). But most consumer motherboards have multiple x16 slots. And I'd rather have a x16 PCIe SSD (theoretically capable of up to 17553MB/s) instead of a x4 PCIe SSD, lol.

(Although, as it turns out, my primary computer is all fulla multi-GPU madness, so my current PCIe Gen2 x8 SSD - serviced off the PCH, not the CPU - fits in quite nicely. Incidentally, it's old SSD tech built around four old SandForce controllers, it achieves high throughput by basically being a hardwired striped RAID of four discrete SSDs working as a single drive on the single card. And it cost my boss way more than I'll comfortably admit, lol.)