Sequential Write

Sequential writes are easier for the SSD controller to process than random writes, so throughput is almost always higher and the bottleneck in this test shifts more to the flash itself. This test puts the most intense load on the flash out of any of our tests, and usually produces the highest overall power consumption. The timing of the sub-tests is the same as for the random tests, and the drive is filled before the test so that features like SLC write caching don't distort the beginning of the test.

Iometer - 128KB Sequential WriteIometer - 128KB Sequential Write (Power)

The heatsink provides a nice boost to the 256GB 950 Pro's sequential write speed, but doubles the sustained performance of the 512GB model. Power efficiency is still improved, especially for the 512GB drive, but it's now reaching almost 6W.

Sequential Write
256GB no heatsink 512GB no heatsink
256GB with heatsink 512GB with heatsink

Without the heatsink, both drives experience thermal throttling only a few minutes into the test, with QD2 performance lower than QD1. With the heatsink, both drives show steady performance across the entire test. The 256GB improves QD1 performance slightly while the 512GB drive sees a 73% increase in QD1 performance. At queue depths greater than one the performance improvements are even larger.

Sequential Read

With the lower overhead of sequential operations and the inherent speed advantage reads have over writes, the sequential read test usually produces the highest throughput numbers. For SATA drives nowadays this usually means the SATA interface itself is the bottleneck, but even with NVMe we haven't seen a drive yet saturate the bandwidth of a PCIe 3.0 x4 link. Depending on the drive, this test could be limited by bottlenecks in the SSD controller's host interface or its connections to the flash memory.

Iometer - 128KB Sequential ReadIometer - 128KB Sequential Read (Power)

Both capacities see significant gains in sequential read performance, accompanied by smaller but still significant increases in power consumption. The 512GB drive once again manages to double sustained performance, and the 256GB improves by 58%.

Sequential Read
256GB no heatsink 512GB no heatsink
256GB with heatsink 512GB with heatsink

Again we see that the drives can't scale with higher queue depths when they're thermally constrained, but when properly cooled they perform superbly with any queue depth larger than one. The 512GB drive manages to deliver slightly more than 2.6GB/s read speed, well above what can be achieved with PCIe 2.0 x4 and about two thirds of the capacity of the PCIe 3.0 x4 link used by the 950 Pro.

Random Performance Mixed Read/Write Performance
Comments Locked

69 Comments

View All Comments

  • FunBunny2 - Monday, December 21, 2015 - link

    ^^
    that's funny
  • Breit - Saturday, December 26, 2015 - link

    You mean something EKWB did for the Intel 750?
    http://www.ekwb.com/news/638/19/EK-is-releasing-In...
    :D
  • meacupla - Monday, December 21, 2015 - link

    I think this is where all those low profile, frag tape backed, RAM heatsinks, that came along with aftermarket GPU coolers, will come in handy.
  • ironwing - Monday, December 21, 2015 - link

    I wonder how much heat 25 LEDs add to a heat sink?
  • mostlyharmless - Monday, December 21, 2015 - link

    Yeah, and aren't the fins supposed to be external instead of internal?
  • Billy Tallis - Monday, December 21, 2015 - link

    They're really more like ducts than fins as designed. What air does manage to enter the heatsink will not all flow across the entire drive. Some of the fresh cool air will bypass part of the drive, and some of it will just be cooling the heatsink itself. It looks like a design that would be very effective with a lot of airflow, but the intake is pretty small.
  • Billy Tallis - Monday, December 21, 2015 - link

    The LEDs draw about 2W. Calculating how much of that gets converted to light that escapes the heatsink is left as an exercise for the reader.
  • MrSpadge - Monday, December 21, 2015 - link

    10 - 20 mW each.
  • r3loaded - Monday, December 21, 2015 - link

    So what we really need is the 950 Pro in a 2.5 inch drive and the U.2 interface. The bonus is drive capacities of 1TB and maybe even 2TB.
  • Impulses - Monday, December 21, 2015 - link

    Great write-up Billy, I appreciate you running the full battery of tests again and re-addressing the Bench scores... A review of a heatsink feels somewhat incomplete without a single temperature measurement tho! SMART readings or something taken with a simple IR thermometer would've been helpful, specially to the kind of DIY'er that's likely to buy this. I'm curious if i can replicate comparable results with a simple stick on heatsink and some airflow, it's particularly relevant for those of us with multiple GPUs where the add in card might put the drive in just as bad of a spot but the M.2 slot might be exposed to more airflow.

Log in

Don't have an account? Sign up now