World's First PCIe 4.0 SSD Demo, Phison E16 Hits 4GB/s

The world of PCIe 4.0 is almost upon us, courtesy of AMD's new third-gen Ryzen 'Matisse' processors, but you can't buy any stupid-fast PCIe 4.0 devices. At least not yet.

Phison looks to change that with its new E16 NVMe SSD controller that it demoed at CES 2019. Phison sells its controllers openly to a wide range of SSD makers, so you'll see these controllers on a plethora of devices in the coming months.

Phison's demo SSD pulled down a blistering 4 GB/s of sequential read throughput and a beastly 4.2 GB/s of write performance during a CrystalBenchMark test. The company is currently using Micron's 96-layer TLC flash, but new E16-powered SSDs will also hit the market with Toshiba's BiCS 4 flash. Phison expects the faster flash to push speeds up to 4.8/4.4 GB/s of read/write throughput.

That will push around big file transfers at a blistering pace, but the E16's random performance, which is most noticeable as snappy responsiveness in the operating system and applications, weighs in north of 900,000 IOPS for both read and write workloads, setting yet another high watermark for performance.

Phison's new PCIe 4.0 x4 SSD controller is so bleeding edge that the company doesn't have a motherboard (that it can show publicly) that supports the faster standard. Instead, the company used a specialized PLDA Gen4HOST adapter card for the demo. The adapter converts the PCIe 3.0 x16 connection to a PCIe 4.0 x8 connection. No fears, though, as Phison has already certified its device as PCIe 4.0-compliant.

The PS5016-E16, as it's formally known, comes with a dual-core ARM controller paired with two of Phison's proprietary CO-X processor cores. The controller has eight addressable channels that can manage four flash packages apiece at 800 MT/s. The demo SSD was equipped with TLC flash, but Phison will also support QLC. Cheaper QLC flash isn't as endurant at TLC, but Phison's LDPC (Low-Density Parity Check) error correction will ensure that endurance is comparable to other devices.

SSDs have already begun to push the limits of the PCIe 3.0 x4 interface, so the step up to PCIe 4.0 will spur a new round of development that pushes SSD controller and flash development to the next level. Not to mention that it will allow you to fling files around at a record pace.

Phison says the SSDs will hit the market in Q3 2019.

Create a new thread in the News comments forum about this subject
25 comments
Comment from the forums
    Your comment
  • Soaptrail
    Does it require an RGB CPU cooler like in the pic above? :)
  • derekullo
    I agree progress is progress, but I was expecting something a bit further from the absolute maximum bandwidth of PCIe 3.0 x4.

    https://en.wikipedia.org/wiki/PCI_Express#History_and_revisions

    The maximum speed of PCIe 3.0 x4 is 3.94 Gigabytes a second.

    The maximum speed of PCIe 4.0 x4 is 7.88 Gigabytes a second.

    PCIe 5.0 x4 which according to Wiki is going to be released in Q2 of 2019, doubles this again to 15.75 Gigabytes a second.

    I was about to ask a question of what is absolute fastest a cpu could transfer data until I realized that the answer lied with the speed of the ram.

    https://en.wikipedia.org/wiki/DDR4_SDRAM#Modules

    This should allow us to conclude a few things ... for example.

    With a PCIe 4.0 X16 device you should be unable to realize its full potential due to its maximum speed being 31.5 Gigabytes a second, but the maximum speed of a respectable DDR4-3200 ram being 25.6 Gigabytes per second.

    This brings up the unexpected and actually kind of funny scenario where you would need to overclock your ram to get the full performance from a PCIe 4.0 x16 SSD adapter.

    https://www.tomshardware.co.uk/aplicata-m.2-nvme-ssd-adapter,review-34022-2.html

    Comes to mind from about a year ago which actually appears to saturate a PCIe 3.0 x8 at 7.88 gigabytes / second at a queue depth of 4 with 10% or so loss to overhead when reading.

    Is this the correct way to think about this?
  • atomicWAR
    Anonymous said:
    I agree progress is progress, but I was expecting something a bit further from the absolute maximum bandwidth of PCIe 3.0 x4.

    https://en.wikipedia.org/wiki/PCI_Express#History_and_revisions

    The maximum speed of PCIe 3.0 x4 is 3.94 Gigabytes a second.

    The maximum speed of PCIe 4.0 x4 is 7.88 Gigabytes a second.

    PCIe 5.0 x4 which according to Wiki is going to be released in Q2 of 2019, doubles this again to 15.75 Gigabytes a second.

    I was about to ask a question of what is absolute fastest a cpu could transfer data until I realized that the answer lied with the speed of the ram.

    https://en.wikipedia.org/wiki/DDR4_SDRAM#Modules

    This should allow us to conclude a few things ... for example.

    With a PCIe 4.0 X16 you should be unable to realize its full potential due to its maximum speed being 31.5 Gigabytes a second, but the maximum speed of a respectable DDR4-3200 ram being 25.6 Gigabytes per second.

    This brings up the unexpected and actually kind of funny scenario where you would need to overclock your ram to get the full performance from a PCIe 4.0 x16 SSD adapter.

    https://www.tomshardware.co.uk/aplicata-m.2-nvme-ssd-adapter,review-34022-2.html

    Comes to mind from about a year ago.

    Is this the correct way to think about this?


    My understanding is running dual channel DDR4 3200 puts you in the 40-50GB/s range depending on you timings. Though I guess you could being going by single channel numbers. My point is that for now we should still have a little wiggle room with PCIe 4.0 vs DDR4 3200 bandwidth but your are correct in the numbers are getting closer. in another few generations an adjustment may need to be made.

    https://www.techspot.com/news/62129-ddr3-vs-ddr4-raw-bandwidth-numbers.html
    https://ark.intel.com/products/126686/Intel-Core-i7-8700-Processor-12M-Cache-up-to-4_60-GHz i7 8700 listed at 41.6GB/s