A 1400 MB/s SSD: ASRock's Z97 Extreme6 And Samsung's XP941

M.2 And SATA Express, Discussed

M.2 PCIe

Intel ratchets up the utility and flexibility of Z97 in a small but profound way. This is achieved by making the chipset’s port 13 and 14 far more flexible than they were in the past. Previously, those two ports facilitated two of the PCH's six SATA interfaces. Now they're flexible, accommodating two big pieces of storage infrastructure: SATA Express and M.2 PCIe.

M.2 PCIe isn’t anything new. As far back as September of last year we were covering that in SanDisk A110 PCIe SSD: Armed With The New M.2 Edge Connector. More recently, we posted Plextor M6e 256 GB PCI Express SSD Review: M.2 For Your Desktop, testing the M.2 PCIe-based SSD on a half-length, half-height add-in board. We were told Plextor was also planning a version without the adapter, which suggested that there'd be motherboards with the corresponding two-lane slots. That day should be today. But as boards equipped with M.2 PCIe slots start selling, SSDs able to drop into them are few and far between.

It's actually pretty easy to mix up M.2 for PCIe and SATA. We're facing the same sort of confusion experienced when mSATA surfaced for mini-PCIe slots. Except this time, it's ever harder to distinguish between M.2 storage with SATA controllers and M.2 PCIe SSDs. So let's just forget about the SATA-based drives and focus on storage natively designed to drop onto the PCI Express bus through the M.2 form factor.

This form factor is flexible in that it can be molded into a wide range of single- or double-sided PCBs. An M.2 device is 22 mm wide, easily fitting a processor and NAND flash packages. Build a longer PCB and you get more space to add flash. Considering Samsung can cram its 1 TB 840 EVO mSATA into roughly the same space as a M.2 2260 (60 mm-long) form factor, something like a M.2 22110 (110 mm-long) gives you a ton of space to work with. And as manufacturing advances, increasing density, it's hard to imagine a day when M.2 will limit the upper bound of capacity.

M.2 Real Estate (in mm)42 mm (M.2 2242)

60 mm (M.2 2260)

80 mm (M.2 2280)
110 mm (M.2 22110)
Single-Sided

924 mm2

1320 mm21760 mm22420 mm2
Double-Sided1848 mm22640 mm23520 mm24840 mm2

Most M.2 PCIe SSDs will utilize two PCI Express lanes (in the case of Z97 Express, of course that means second-gen transfer rates). But Samsung's XP941 is unique in that it communicates over four. That makes it the most ideal candidate for testing ASRock's Z97 Extreme6 and its four-lane PCI Express 30 Ultra M.2 socket.

SATA Express

SATA Express replaces SATA 6Gb/s. The Serial ATA International Organization realized that doubling SATA's transfer rate again wasn't going to be practical. As Paul Wassenberg of Marvell fame told me last year at Flash Memory Summit 2013, SATA Express makes a lot more sense.

As the working group scaled up SSDs from one to eight PCIe lanes in testing, power consumption went through the roof as lanes were attached. But with just two lanes at third-gen transfer rates, power didn’t increase much compared to a SATA 6Gb/s-connected equivalent drive, even as performance was vastly superior. We know from the challenges presented by 12 Gb/s SAS that a cost-effective implementation would be difficult to achieve for SATA. Meanwhile, SATA Express wouldn't be as problematic. Given its PCI Express roots, however, cabling was the challenge to address.

Unlike M.2 PCIe SSDs, which can span up to four lanes, SATA Express uses just two. But whereas a M.2 PCIe-based drive is basically stuck to the motherboard, SATA Express employs cables to make more remote connections, just like SATA. This poses a few practical issues. An external PCIe-based SSD needs a signal from the clock generator. Carrying that signal over distance requires shielding and a beefier (more expensive) cable. So, to combat a prohibitively pricey implementation, the signal can be provided by the solid-state device itself.

Gain a Port, Lose a Port

Based on Intel's implementation of SATA Express in Z97, if you utilize the new technology, you lose access to two of the storage controller's SATA 6Gb/s ports and the M.2 interface. If you instead choose to go with M.2 (the devices are more plentiful, after all), you can't use SATA Express.

To help add a bit of clarity, I created the flow chart above to clarify M.2 PCIe and SATA Express, along with the AHCI and NVMe interface specifications.

On Z97, the PCH-provided M.2 and SATA Express ports are mutually exclusive. You cannot use both simultaneously. Asus is adding third-party SATA Express controllers to some boards, so obviously those are able to operate independently. And then there's ASRock's solution: borrowing four lanes from the CPU's PCI Express controller to create the Ultra M.2 slot. Let's look at that in more depth...

Create a new thread in the US Reviews comments forum about this subject
This thread is closed for comments
20 comments
    Your comment
  • aminebouhafs
    Once an SSD in plugged into the Ultra M.2 slot, the bandwidth between central processing unit and graphics processing unit is cut-down by half. Therefore, while the end-user gets additional SSD performance, the end-user may lose some GPU performance because of insufficient bandwidth between it and the CPU.
    -8
  • JoeArchitect
    Very interesting article and a great read. Thanks, Chris - I hope to see more like this soon!
    8
  • wussupi83
    great article! - although z97 still seems boring
    0
  • Eggz
    This makes me excited for X99! With 40 (or more) lanes, of PCI-e (probably more), there will be no need to compromise. We have to remember that the Z97 Chipset is a consumer-grade product, so there almost has to be tradoffs in order to justify stepping up to a high-end platform.

    That said, I feel like X99, NVMe, and and M.2 products will coincide nicely with their respective releases dates. Another interesting piece to the puzzle will be DDR4. Will the new storage technology and next-generation CPUs utilize it's speed, or like DD3, will it take several generations for other technologies to catch up to RAM speeds? This is quite an interesting time :)
    -1
  • Amdlova
    Chris test the asrock z97 itx... and another thing... my last 3 motherboard from asrock and i want to say Asrock Rock's!
    5
  • Damn_Rookie
    While storage isn't the most important area of computer hardware for me, I always enjoy reading Christopher's articles. Very well written, detail orientated, and above all else, interesting. Thanks!
    0
  • hotwire_downunder
    ASRock has come along way, I used them a long time back with disappointing results, but I have started to use them again and have not been disappointed this time around.

    Way to turn things around ASRock! Cheap as chips and rock steady!
    0
  • alidan
    @aminebouhafs if i remember right, didn't toms show how much performance loss there is when you tape gpu cards to emulate having half or even a quarter of the bandwidth? if i remember right back than the difference was only about 12% from 16 lanes down to either 4 or 8
    0
  • Eggz
    269694 said:
    @aminebouhafs if i remember right, didn't toms show how much performance loss there is when you tape gpu cards to emulate having half or even a quarter of the bandwidth? if i remember right back than the difference was only about 12% from 16 lanes down to either 4 or 8


    PCI-e 3.0 x8 has enough bandwidth for any single card. The only downside to using PCI-e lanes on the SSD applies only to people who want to use multiple GPUs.

    Still, though, this is just the mid-range platform anyway. People looking for lots of expansion end up buying the X chipsets rather than the Z chipsets because of the greater expandability. I feel like the complaint is really misplaced for Z chipsets, since they only have 16 PCI-e lanes to begin with.
    1
  • cryan
    Quote:
    Once an SSD in plugged into the Ultra M.2 slot, the bandwidth between central processing unit and graphics processing unit is cut-down by half. Therefore, while the end-user gets additional SSD performance, the end-user may lose some GPU performance because of insufficient bandwidth between it and the CPU.


    Well, it'll definitely negate some GPU configurations, same as any PCIe add-in over the CPU's lanes. With so few lanes to work with on Intel's mainstream platforms, butting heads is inevitable.

    Regards,
    Christopher Ryan
    2
  • cryan
    Quote:
    While storage isn't the most important area of computer hardware for me, I always enjoy reading Christopher's articles. Very well written, detail orientated, and above all else, interesting. Thanks!


    Awww, shucks!

    Regards,
    Christopher Ryan
    3
  • obamaliar
    Supercool review, Just as a note though any pair of good SATA Based SSD's will blow the doors off of that x941. For example I am getting 740.00MB/s bandwidth at steady 5 and 948.37MB/s at recovery 5 for PCM8 extended photoshop heavy from a pair of Intel 730's
    -1
  • obamaliar
    Supercool review, Just as a note though any pair of good SATA Based SSD's will blow the doors off of that x941. For example I am getting 740.00MB/s bandwidth at steady 5 and 948.37MB/s at recovery 5 for PCM8 extended photoshop heavy from a pair of Intel 730's
    -1
  • Evolution2001
    obamaliar, how do you reckon that your 740MBps or 948MBps is faster than 1400MBps? (referencing the sequential read of the tested drive)
    SATA3 has a theoretical max of 6Gbps (750MBps). However, the practical max is more around 600MBps.
    Assuming you are running your Intel 730's in RAID-0 and achieving the max practical throughput, you'd still only come up with ~1200MBps which is slower than what Tom's saw at 1400MBps ON A SINGLE DRIVE.
    1
  • obamaliar
    Quote:
    obamaliar, how do you reckon that your 740MBps or 948MBps is faster than 1400MBps? (referencing the sequential read of the tested drive) SATA3 has a theoretical max of 6Gbps (750MBps). However, the practical max is more around 600MBps. Assuming you are running your Intel 730's in RAID-0 and achieving the max practical throughput, you'd still only come up with ~1200MBps which is slower than what Tom's saw at 1400MBps ON A SINGLE DRIVE.

    Quote:
    obamaliar, how do you reckon that your 740MBps or 948MBps is faster than 1400MBps? (referencing the sequential read of the tested drive) SATA3 has a theoretical max of 6Gbps (750MBps). However, the practical max is more around 600MBps. Assuming you are running your Intel 730's in RAID-0 and achieving the max practical throughput, you'd still only come up with ~1200MBps which is slower than what Tom's saw at 1400MBps ON A SINGLE DRIVE.
    Evolution 2001, I am referring to OS simulated performance IE CRYAN's PCMark 8 extended testing. Please read the article and you would understand that sequential performance is really a non-factor in comparison to random performance in an OS environment. Right now, SATA RAID has vastly superior random performance to PCIe drives like the X941, even if you were to soft RAID a pair of X941's together they cannot match a pair of good SATA SSD's in RAID in an OS environment. I cannot show you that exactly because Soft RAID is not bootable. look here: https://www.facebook.com/groups/1445011539065390/ there ia a pair of X941's soft raided getting their asses kicked by SATA RAID. The reason? 4K writes do not scale on PCIe drives. When PCIe drives can be RAIDed, bootable in RAID and have an RST type driver that allows for write caching Then they will become the superior OS disk.
    -1
  • cryan
    303076 said:
    Quote:
    obamaliar, how do you reckon that your 740MBps or 948MBps is faster than 1400MBps? (referencing the sequential read of the tested drive) SATA3 has a theoretical max of 6Gbps (750MBps). However, the practical max is more around 600MBps. Assuming you are running your Intel 730's in RAID-0 and achieving the max practical throughput, you'd still only come up with ~1200MBps which is slower than what Tom's saw at 1400MBps ON A SINGLE DRIVE.
    Quote:
    obamaliar, how do you reckon that your 740MBps or 948MBps is faster than 1400MBps? (referencing the sequential read of the tested drive) SATA3 has a theoretical max of 6Gbps (750MBps). However, the practical max is more around 600MBps. Assuming you are running your Intel 730's in RAID-0 and achieving the max practical throughput, you'd still only come up with ~1200MBps which is slower than what Tom's saw at 1400MBps ON A SINGLE DRIVE.
    Evolution 2001, I am referring to OS simulated performance IE CRYAN's PCMark 8 extended testing. Please read the article and you would understand that sequential performance is really a non-factor in comparison to random performance in an OS environment. Right now, SATA RAID has vastly superior random performance to PCIe drives like the X941, even if you were to soft RAID a pair of X941's together they cannot match a pair of good SATA SSD's in RAID in an OS environment. I cannot show you that exactly because Soft RAID is not bootable. look here: https://www.facebook.com/groups/1445011539065390/ there ia a pair of X941's soft raided getting their asses kicked by SATA RAID. The reason? 4K writes do not scale on PCIe drives. When PCIe drives can be RAIDed, bootable in RAID and have an RST type driver that allows for write caching Then they will become the superior OS disk.


    Actually, the 4 KB writes are really an artifact of the AHCI controller/API. If you took the same flash and controller on the Sammy, but rigged it to use NVMe, I think you'd see a big bump in random 4 KB performance. I've said over and over that desktop users, for now, are better off by using a couple SATA drives in RAID. More than just adding bandwidth, which isn't always important (strictly speaking), it lowers service times significantly. Plus, it's great to just keep adding cheap drives and getting more performance and capacity (when striped). See the Plextor M6e PCIe review for my thoughts on this.

    It's all academic anyway, since you can only buy the XP941 from a few random places, and it's $750. If I had a laptop which could use it, maybe I go that route, but even there SATA is just more power efficient. Give me a 1 TB EVO or M550 instead..... at least for the time being.

    PS: Is this Jon C??

    Regards,
    Christopher Ryan
    2
  • obamaliar
    Thanks for the reply love your stuff C Ryan :)
    0
  • Eggz
    1302170 said:
    Give me [either a 750 GB] or a 1 TB EVO or M550 instead..... at least for the time being.


    Totally agree! For now.

    I also added the 750 EVO in there because (I believe) the only difference between the 1TB and the 750GB is capacity, unlike the smaller drives, which actually have less performance (i.e. 120, 250, & 500 GB).
    1
  • logainofhades
    I would rather use a single powerful GPU anyway, so the cut to 8x due to the ultra M.2 slot doesn't bother me at all. This is definitely and interesting board. I want an Ultra M.2 slot on a mini-itx board. :D
    0
  • lukebutters
    If the RAID controller is set up with 4 disks in RAID 10, will the DMI limit be reached?
    0