Internals, Testing Methodology and System Setup




Front (480GB top, 280GB bottom):

Controller close-up:

This is Intel's 7-channel controller design that has been around since the SSD DC P3700/SSD 750 (though it's 14-channels in those parts – same PCB layout though). We suspect significant changes had to be made to bring latencies down so significantly. Also note: no DRAM. Who needs DRAM with a few hundred GB of stuff that is almost as fast!

Rear (480GB top, 280GB bottom):

Chip counts and analysis:

16GB and 32GB 3D XPoint packages side by side (480GB model).

  • 280GB: 14x16GB front / 7x16GB rear – 336GB total
  • 480GB: 7x16GB+7x32GB front / 14x16GB rear – 560GB total

Now I know what you're thinking. If Optane doesn't need overprovisioning or a large spare area in the same sense as NAND flash SSDs, then why on earth do these parts have on the order of 20% extra 3D XPoint installed? Intel's official response to that question is “3D XPoint™ memory media does not require over-provisioning.  We aren’t disclosing the use of the additional media at this time.”, which is fair enough, but I can hypothesize a bit based on my knowledge of the history of this controller in particular and of storage media technologies in general:

  • Raw 3D XPoint dies do not have built-in space or mechanisms for ECC (error correction).
  • Remember these are addressed more like RAM than like NAND flash. It is entirely possible that Intel set up their channels in a RAID-6 style configuration, effectively making it into a 12-channel controller with two effective channels for parity.
  • After doing that math, we still end up with ~3.3% left over for any metadata or spares.
  • Unproven, newer technologies will typically be made with a greater design margin.

Kudos for Intel for (theoretically) going way over the top on error correction and redundancy here. It makes sense to do this for the first iterations of any storage product, regardless of the companies confidence in the technology.

Testing Methodology

Our tests are a mix of synthetic and real-world benchmarks. IOMeter, HDTach, HDTune, Yapt and our custom File Copy test round out the selection to cover just about all bases. We have developed a custom test suite as off-the-shelf tests just no longer cut it for in-depth storage testing. More details on the next page. If you have any questions about our tests just drop into the Storage Forum and we'll help you out!

Test System Setup

We have several storage testbeds. A newer ASUS P8Z77-V Pro/Thunderbolt and a Gigabyte Z170X SOC Force (for RAID testing). Future PCIe and SATA device testing, including this review, take place on an ASUS Sabertooth X99, which comes equipped with USB 3.1, M.2, and can also handle SFF-8639 (U.2) devices with the proper adapter.

PC Perspective would like to thank Intel, ASUS, Gigabyte, Corsair, Kingston, and EVGA for supplying some of the components of our test rigs. 

Hard Drive Test System Setup
CPU Intel Core i7 5820K @ 4.125 GHz
Motherboard ASUS Sabertooth X99
Memory 16GB Micron DDR4 @ 3333
Hard Drive G.Skill 32GB SLC SSD
Sound Card N/A
Video Card EVGA GeForce GTX 750
Video Drivers GeForce Game Ready Driver 347.88
Power Supply Corsair CMPSU-650TX
DirectX Version N/A
Operating System Windows 8.1 Pro X64 (update)
  • PCPer File Copy Test
  • HDTach
  • HDTune
  • IOMeter
  • YAPT
« PreviousNext »