1. Forum
  2. Support
  3. Customer Support
  4. SB-RKT4P-4TB slower read speeds in linux over time

SB-RKT4P-4TB slower read speeds in linux over time

  • l

    I have two SB-RKT4P-4TB SSDs running on a Linux system with kernel 6.2.0-37-generic

    The speed was originally fine on them a couple of years ago, but the drives have been getting noticebly slower and slower over time.  Usage patterns have not changed.  Originally when I started/opened OpenOffice for the first time it would display the initial splash-dialog in less than a second, but now it takes a solid 5-10 seconds.

    Smartctl yields no errors.  Trim is enabled and has been running weekly - most recent result:

    Oct 06 00:03:42 mint2 fstrim[2657178]: /: 80.2 GiB (86149525504 bytes) trimmed on /dev/nvme0n1p2

    Used fio for some randread tests on them:

    fio --name=read_iops --directory=. --size=10G \
    --time_based --runtime=20s --ramp_time=2s --ioengine=libaio --direct=1 \
    --verify=0 --bs=4K --iodepth=64 --rw=randread --group_reporting=1

    results are very similar between the two:

    read: IOPS=349k, BW=1362MiB/s (1428MB/s)(26.6GiB/20001msec)

    read: IOPS=342k, BW=1337MiB/s (1402MB/s)(26.1GiB/20001msec)

    My understanding is that these drives should be much closer to 6000MB/s  rather than 1400MB/s

    Please advise

     

    0 likes
  • TS

    Hello @linuxssd, 

    The fio test you ran measured small-block random read performance (4K blocks, high queue depth). These results are normal for that workload and do not reflect the drive’s maximum sequential read throughput. To evaluate the expected performance, you’ll need to run a large-block sequential read test. 

    Regarding your application slowdown issue, I will check in with the team and let you know if I can find a resolution. 

  • l

    Thank you for replying.  I should point out that the OpenOffice scenario was just a relatable example.  Any operation that has to access many small files has grown much slower over time.  Another example is simply "grep"ing recursively a directory of thousands of files.  Similarly, when receiving a new version of a particular library from a vendor and sync'ing it in Intellij Idea (a java IDE) it has to unzip/scan through thousands of new files and that takes much longer now also.  There are many more examples but the common theme appears to be reading lots of small files.   That's why I thought randread would be a better test than sequential.   Is there a different test I can run to identify what the cause of the slowdown is?  Or is there something I can measure while I'm doing one of these slow activities that will yield insights into the root cause?

    Additionally, the usage of these drives is only at 44% and 29% (i.e. to rule-out the scenario where SSDs drop in performance when they are closer to 100% capacity)

    0 likes
  • l

    I'm reading about TLC NAND degradation over time and came across this bit of information:

    Read-retry optimization: As cells degrade, the SSD may need to re-read a page multiple times with adjusted voltage levels to retrieve data correctly, which increases latency.

    Is there a test I can run on these SSDs to determine how many times they might be re-reading a page until the data is retrieved correctly? 

    0 likes
  • TS

    @linuxssd 

    Thank you for the additional details and examples. The slowdown on many small-file operations aligns with some behaviors we’ve encountered related to how the controller manages older data and read retries over time.

    Could you please provide the serial # (on the backside label) and the current firmware revision of your drive? This will help determine if you are on the version where we expect such behavior or if a firmware update might help. We appreciate your patience as we investigate.

    0 likes
  • l

    drive 1 - 470207270EA300012025, firmware R4PB47.3

    drive 2 - 470207270EA300012021, firmware R4PB47.3

    On another note, I am not receiving emails when someone replies on this thread even though the setting is enabled on my profile.  Is there a different setting I need to change in order to be notified when someone replies?

    Thanks much

    0 likes
  • l

    @The Sabrent Informer

    Checking back to see if you saw my reply where I provided the details you requested.  Any update?

    Thanks

    0 likes
  • S

    @linuxssd Sorry about that. You can probably get the firmware update through support with the details, if not there's a link here. As for email notification, this is currently not operational but thank you for making us aware of it. It's something we can fix.

    0 likes
  • l

    @Sabrent @The Sabrent Informer Please note that the other representative stated the following:

    The slowdown on many small-file operations aligns with some behaviors we’ve encountered related to how the controller manages older data and read retries over time.

    They also stated that they could determine that from my serial numbers.  Are my units affected by this problem?  If so, will you please confirm that a firmware update would actually fix it?  Updating the firmware is going to be a huge ordeal and I'd rather not do it unless it will actually work.

    Thanks

    0 likes
  • S

    @linuxssd Yes, this firmware should fix the issue.

    0 likes
  • M

    The linked firmware tool doesn't work, it throws an error upon running it for the first time (when it extracts multiple files from the exe). Looks like it is missing a pcie_dll.dll file, which from what I can tell is supposed to be something from Phison.

    If I ignore this error and continue, the program runs, but I think it might be running strangely due to the missing file. It shows only a single option to "reinitial" under "Function". Not sure if this means to do a firmware upgrade or some other more dangerous operation that could potentially wipe the data on the drive. Well, I have the data backed up on mine, so i bravely tried the "reinitial" operation anyways. It didn't work and failed with the message "load burner fail". The utility would also sometimes BSOD crash Windows with a DPC_WATCHDOG_VIOLATION error. Screenshots attached.

     

    0 likes
  • l

    @Sabrent @The Sabrent Informer 

    It's very strange to post a dropbox link for firmware updates and I do not trust external links.  Why is this not available on the Sabrent website?  The only firmware update/download available on Sabrent website is:


    https://downloads.sabrent.com/product/sb-rkt4p-xxx-firmware-upgrade/?wpdmdl=422&refresh=624f7ccbe7dd51649376459&ind=1649359183176&filename=SB-RKT4P-XXX%20Firmware%20Update.zip


    However, it only says that it updates firmware from 1.1 to 1.2.  My firmware version is R4PB47.3.  Does that equate to 1.1 or 1.2 or some other version?  Is 1.2 the newest firmware?  Where are the patch/release notes for firmware 1.2 so I can review what changed for applicability?


    Can you please provide more details on the nature of the problem I'm experiencing and how specifically it is addressed by the 1.2 firmware?  The reports by @McChen  of the firmware update errors make me even more trepedatious about attempting it.  Can someone speak to the firmware errors and how to resolve?  (Thank you @McChen  for trying this out - are you having the same problem as me?)

    0 likes
  • M

    Sabrent does post some files on dropbox, some downloads even from the official website point to a dropbox file. I do believe the file is authentic, but I get why it would arouse suspicion.

    There are multiple hardware revisions of the Rocket 4 Plus SSD, at least 3 that I have seen. The firmware posted on the website that is v1.2 is only for the first revision that used Micron B27B flash chips. The version you and I have is the second version that upgraded to faster Micron B47R flash and uses a different R4PB47.x firmware. I also have a newer 3rd version that uses a R184PL.x firmware, but there is little information about what components this version uses. I believe it is also an E18 controller, but probably different NAND.

    I believe this firmware likely addresses a known issue with the Phison E18 SSD controller that all the Rocket 4 Plus revisions use where performance degrades over time. Files that were stored on the drive were not being refreshed correctly in the background and thus the read speeds and read latency would degrade significantly over time. Phison has issued fixed firmware and it has been slowly trickling out to all their OEM partners. I'm not sure if if it affects every drive that uses the E18 controller or only certain drive configurations. I have not really noticed it on mine yet, since I use my Rocket drive primarily for large size backups and the files get overwritten frequently. I don't do a lot of reading from the drive either, mostly writes (which the Rocket 4 Plus v2 is really great at, long sustained write performance).

    Would be great to get Sabrent's confirmation of what the new firmware fixes though. Sabrent, is there also a similar update for the R184PL.x revision that I have?

    0 likes
Please login to reply this topic!

Search

Forum statistics

Page views (24h):
666
Page views (30d):
16,608
Topics:
331
Posts:
1,382
Members:
2,448

Newest members

© 2025, Sabrent | Powered by Xenforum