AnandTech DAS Suite - Benchmarking for Performance Consistency

Our testing methodology for DAS units takes into consideration the usual use-case for such devices. The most common usage scenario is transfer of large amounts of photos and videos to and from the unit. Other usage scenarios include the use of the DAS as a download or install location for games and importing files directly off the DAS into a multimedia editing program such as Adobe Photoshop. Some users may even opt to boot an OS off an external storage device.

The AnandTech DAS Suite tackles the first use-case. The evaluation involves processing three different workloads:

  • Photos: 15.6 GB collection of 4320 photos (RAW as well as JPEGs) in 61 sub-folders
  • Videos: 16.1 GB collection of 244 videos (MP4 as well as MOVs) in 6 sub-folders
  • BR: 10.7 GB Blu-ray folder structure of the IDT Benchmark Blu-ray

Each workload's data set is first placed in a 25GB RAM drive, and a robocopy command is issued to transfer it to the DAS under test (formatted in NTFS). Upon completion of the transfer (write test), the contents from the DAS are read back into the RAM drive (read test). This process is repeated three times for each workload. Read and write speeds, as well as the time taken to complete each pass are recorded. Bandwidth for each data set is computed as the average of all three passes.

 

Blu-ray Folder Read

Despite the presentation of all the results for a given workload in one graph, the numbers above require careful analysis - we are essentially looking at three different sets:

  1. Thunderbolt 3 SSDs with JHL7540 host
  2. Thunderbolt 3 SSDs with JHL6540 host
  3. USB 3.2 Gen 2 SSDs

Only the first two sets are perfectly apples-to-apples - in fact, that is the case only when we consider the 2TB drives. The 4TB variant stands alone, as we have not evaluated any other 4TB Thunderbolt 3 SSD earlier. On the USB side, out evaluations so far have used the JHL6540's USB host capabilities - but the XTRM-Q variants could only be evaluated in this mode using the ASMedia ASM2142.

Going back to the results, we see our DIY SSD coming out on top across most workloads with the JHL7540 host. A notable exception is for the video reads, where the 2TB XTRM-Q wins out. However, it must be kept in mind that we are comparing 1TB, 2TB, and 4TB drives in this set. Moving on to the JHL6540 host, we again see the write workloads turning out to be the weak point for the XTRM-Q drives. The 2TB Plugable and OWC drives are apples-to-apples and they consistently out-perform (or match, at worst) the XTRM-Q. The usage of QLC NAND is likely to be the culprit in this case. In the USB mode, the XTRM-Q drives acquit themselves well. There is no significant gulf in the numbers between the different USB 3.2 Gen 2 SSDs in this mode. For all practical purposes, the casual user will notice no difference between them in the course of normal usage.

Performance Consistency

Beyond basic file copying benchmarks, power users may also want to dig deeper to understand the limits of each device. To address this concern, we also instrumented our evaluation scheme for determining performance consistency. Aspects influencing the performance consistency include SLC caching and thermal throttling / firmware caps on access rates to avoid overheating. This is important for power users, as the last thing that they want to see when copying over 100s of GB of data is the transfer rate going down to USB 2.0 speeds.

In addition to tracking the instantaneous read and write speeds of the DAS when processing the AnandTech DAS Suite, the temperature of the drive was also recorded at the beginning and end of the processing. In earlier reviews, we used to track the temperature all through. However, we have observed that SMART read-outs for the temperature in NVMe SSDs using USB 3.2 Gen 2 bridge chips end up negatively affecting the actual transfer rates. To avoid this problem, we have restricted ourselves to recording the temperature at either end of the actual workloads set. The graphs below present the recorded data.

Performance Consistency and Thermal Characteristics

The first three sets of writes and reads correspond to the photos suite. A small gap (for the transfer of the video suite from the internal SSD to the RAM drive) is followed by three sets for the video suite. Another small RAM-drive transfer gap is followed by three sets for the Blu-ray folder. An important point to note here is that each of the first three blue and green areas correspond to 15.6 GB of writes and reads respectively. The low write speeds with Thunderbolt 3 hosts are evident in the graphs, but they are remarkably consistent across different sets of workloads pointing to the SLC cache not running out for this traffic pattern. The thermal design is also excellent - the internal SSD temperaturee goes slightly north of 50C in only one of the six graphs of interest above.

Synthetic Benchmarks - ATTO and CrystalDiskMark PCMark 10 Storage Bench - Real-World Access Traces
POST A COMMENT

14 Comments

View All Comments

  • ballsystemlord - Monday, December 14, 2020 - link

    @Ganesh When selecting "print article" the ATTO, and PCMark10 benchmarks don't show up. There might be others missing. I didn't do a page for page comparison. Reply
  • ganeshts - Monday, December 14, 2020 - link

    Though I can't do much in terms of a solution for this, I can give you the technical explanation:

    The 'Print View' unfortunately strips out everything between <script> and </script> on a page - so, none of the drop-downs actually work in that view. In addition, this behavior makes pages with multiple script tags for different drop-downs end up with the content between them going kaput (as you have noted).
    Reply
  • RSAUser - Tuesday, December 15, 2020 - link

    In Firefox: right click, inspect element, look for the div with a class of articleContent and hit right click on it and select screenshot node. Reply
  • ricebunny - Monday, December 14, 2020 - link

    I noticed the issue of SLC write caching not enabled when trying out the Orico TB3 aluminum enclosure. Both the 970 Evo Plus and Kingston KC2500 NVMe drives topped out at only 880MB/s write in CrystalMark.

    I think this warrants further investigation.
    Reply
  • Tomatotech - Monday, December 14, 2020 - link

    Is there a way to buy just the chipset - maybe on Alibaba? Reply
  • Deicidium369 - Wednesday, December 16, 2020 - link

    You mean the Intel controllers? Are you wanting to buy more than 10,000? Reply
  • hubick - Monday, December 14, 2020 - link

    "a unique product line in the market – a dual-mode SSD that can work with both Thunderbolt 3 ports and USB ports"

    Not entirely unique - I'm typing this on a Linux system booted off a 2TB "LaCie Rugged SSD Pro" which uses the same JHL7440 Titan Ridge chipset.
    Reply
  • ganeshts - Monday, December 14, 2020 - link

    Does it work when connected to a USB host? As far as I could find in my research, this is the only SSD that works when connected to a USB host and also a Thunderbolt 3 host *with higher performance* in the latter case. Reply
  • hubick - Monday, December 14, 2020 - link

    Yes. I'm actually swapping between USB on this Threadripper desktop, my Hades Canyon NUC, and soon a Razer Book 13 if it ever gets here :-) Reply
  • Deicidium369 - Wednesday, December 16, 2020 - link

    I have looked as well - tons of Orico HD docking.... Maybe he could post the model number. Reply

Log in

Don't have an account? Sign up now