Performance Consistency

Our performance consistency test explores the extent to which a drive can reliably sustain performance during a long-duration random write test. Specifications for consumer drives typically list peak performance numbers only attainable in ideal conditions. The performance in a worst-case scenario can be drastically different as over the course of a long test drives can run out of spare area, have to start performing garbage collection, and sometimes even reach power or thermal limits.

In addition to an overall decline in performance, a long test can show patterns in how performance varies on shorter timescales. Some drives will exhibit very little variance in performance from second to second, while others will show massive drops in performance during each garbage collection cycle but otherwise maintain good performance, and others show constantly wide variance. If a drive periodically slows to hard drive levels of performance, it may feel slow to use even if its overall average performance is very high.

To maximally stress the drive's controller and force it to perform garbage collection and wear leveling, this test conducts 4kB random writes with a queue depth of 32. The drive is filled before the start of the test, and the test duration is one hour. Any spare area will be exhausted early in the test and by the end of the hour even the largest drives with the most overprovisioning will have reached a steady state. We use the last 400 seconds of the test to score the drive both on steady-state average writes per second and on its performance divided by the standard deviation.

Steady-State 4KB Random Write Performance

No, that's not a mistake. The abysmal steady-state performance of both MK8115 prototypes is a consequence of not having a cache for the logical to physical address mapping. Each write in this test is overwriting an in-use logical block address. While the controller's wear leveling ensures that the new data will go to a new location (the physical address of which is probably kept in a register on the controller), the controller has to read from the flash to figure out which physical page just got invalidated and became a candidate for garbage collection. Meanwhile, the garbage collection process has to scan the flash instead of a DRAM cache in order to determine if all the pages in an erase block contain stale data or if there's some live data that needs to be moved before the controller can erase that block.

This test isn't easy for any consumer SSD, but a DRAM-less SSD suffers acutely. Still, in spite of all the barriers to efficient flash management, the MK8115 drives are faster than a hard drive.

Steady-State 4KB Random Write Consistency

The MK8115 samples have poor steady-state throughput, but their consistency is pretty good. Maxiotek has ensured that the garbage collection never completely stalls the drive's progress on completing host write operations.

IOPS over time
Default
25% Over-Provisioning

Before the MK8115 SSDs run out of spare area, their random write speeds aren't horrific, and are better than what most SSDs manage in steady state. It's only when the spare area runs out that the MK8115 gets into serious trouble. The TLC drive takes longer to reach this point, because the odd 48GB capacity of Micron's 3D TLC forced Maxiotek to give the drive much more overprovisioning than a typical 512GB SSD.

Steady-State IOPS over time
Default
25% Over-Provisioning

There is some cyclical behavior in the steady state of the MK8115 drives, but less than for most SSDs. The MLC drive in particular is extremely consistent save for periodic but very short bursts of higher performance.

Inroduction AnandTech Storage Bench - The Destroyer
Comments Locked

60 Comments

View All Comments

  • MajGenRelativity - Tuesday, May 9, 2017 - link

    same
  • vladx - Tuesday, May 9, 2017 - link

    It will take until Q2 2018 for prices to start going down again.
  • milli - Tuesday, May 9, 2017 - link

    You should have included other DRAM-less drives.
  • Billy Tallis - Tuesday, May 9, 2017 - link

    I would have, if I had any on hand to test. Nobody's offered one up for review in quite a while, and I've asked a few vendors for a Phison S11 drive but they're not interested.
  • vladx - Tuesday, May 9, 2017 - link

    It's pretty obvious why, at least Maxiotek has some guts.
  • bortiz - Tuesday, May 9, 2017 - link

    My problem with dram-less ssd is the durability of the drive. How does the life-expectancy of this drive compare to other SSD drives. The DRAM in current SSD drives groups data accesses together, reducing the data array access. This increases performance and array durability. Array durability is proportional to the number of array accesses. I don't think this is a good idea and I would like to see if someone can prove to me that this will have a decent life span (5 years???)
  • vladx - Tuesday, May 9, 2017 - link

    Whether a SSD IS DRAM-less or not doesn't influence life expectancy.
  • Billy Tallis - Tuesday, May 9, 2017 - link

    SSDs that have external DRAM usually don't use it as a write cache for user data; it's just for caching the mapping metadata. A DRAMless controller like MK8115 can do pretty much the same amount of write combining and wear leveling that a mainstream controller can. There may be a slight increase in write amplification from the controller having to flush metadata updates to the flash more often, but mainstream drives don't want to buffer those writes too much either, for the same safety reasons they don't put user data in the DRAM.

    Remember that write operations are the only ones that significantly affect drive lifetime. DRAMless controllers need to do more flash reads, but read disturb errors are still too rare for that to matter to drive reliability.
  • CrazyElf - Tuesday, May 9, 2017 - link

    It is difficult for it to make the cut with a low 4k random read and write test like this SSD has.

    I think that the DRAM cache is well worth it.

    I think it is best to hold off on SSDs for now.
  • rocky12345 - Tuesday, May 9, 2017 - link

    Yea no thanks if this is what DRAM-less drives are like I think I will stick to SSD's that can maintain their performance a lot better. I don't care if these drives are a bit cheaper if you end up swearing at your computer because the SSD is lagging then for peace of mind just spend the extra cash on a real SSD.

    I just sold a gaming system to a customer that had a SSD installed as the windows drive. Yep it booted fast and was pretty peppy but if you tried to work the system it seemed like the SSD would bog right down to a crawl so I would assume it was a DRAM-Less SSD. When I sold the customer on the system I was going off of my own SSD usage on my gaming laptop & gaming desktop options which would never bog down like that. I guess the good news was the customer did come back a few days later and said the system was great and so fast that he never had to wait for stuff to load like he did before on his older setup so he was a happy camper I guess.

Log in

Don't have an account? Sign up now