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

The SM2260 sample has relatively poor steady-state random write performance given that it uses NVMe and MLC NAND. The use of SLC caching and the lower than normal spare area of this drive both contribute to poor steady-state performance but may not significantly impair short-term performance.

Steady-State 4KB Random Write Consistency

The SM2260 sample has an even lower consistency score than the Intel SSD 600p, which uses basically the same controller and TLC NAND, but has substantially more spare area.

IOPS over time
Default
25% Over-Provisioning

After the very short initial burst of great performance around 140k IOPS, the SM2260 sample transitions abruptly to a steady state that it maintains throughout the rest of the test with no long-term shifts in behavior. With extra overprovisioning reserved, there's an intermediate phase consisting of mostly performance around 80k IOPS and second burst at 140k IOPS before a higher performance but no more consistent steady state is reached.

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

Looking more closely at the steady state, the SM2260 sample is mostly shifting between four performance levels, with the most common being around 5k IOPS. However, when it goes through periodic phases of lower performance, it is stuttering hard and will often go for an entire second without completing any I/O. This is clearly poorly-managed garbage collection, possibly exacerbated by thermal throttling and definitely suffering from insufficient spare area.

With more overprovisioning, the severe stuttering is all but eliminated and the normal performance range jumps to around 24k IOPS with periods where it drops to around 6k IOPS.

Introduction AnandTech Storage Bench - The Destroyer
Comments Locked

27 Comments

View All Comments

  • motigez - Sunday, February 19, 2017 - link

    Honestly, at this point, TLC is the only viable technology for Client market, the incremental performance benefit you may get on MLC based SSD just don't worth it, and I expect to see little to no traction on those moving forward.
  • LordanSS - Sunday, February 19, 2017 - link

    To be very honest, a 3D MLC at 40nm has very good endurance, which would be good for a usage of heavy writes.

    If you're dealing with video capture and editing, it'd be a good choice over 3D TLC, even tho 3D TLC at 40nm has decent endurance as well.
  • leexgx - Tuesday, February 21, 2017 - link

    to bad they did not continue the BX100 line as it had insane low power use (still fast enough for a low end SSD) BX200 and higher was like the worst SSD for low end SSD and power use
  • Ranger1065 - Wednesday, February 22, 2017 - link

    Does anyone really care about this? The number of comments seems to indicate a comprehensive "NO." Another article, another step towards oblivion. Great job Anandtech.
  • BrokenCrayons - Wednesday, February 22, 2017 - link

    I see you're back to once again try to decry the site. It's funny if things are so bad that you keep coming back to read articles, but that's a special sort of problem, I suppose. Anyway, since you're having trouble with reading comprehension, most of the comments swirl around the drive's poor performance relative to other products. Just because someone doesn't leap to their feet to applaud a device that landed at the bottom of a chart doesn't mean the reviewer, the article, or the company publishing it is doing something wrong by reporting their test results. Just because people are generally in agreement with the results via their comments certainly doesn't indicate there's a problem with the review. You're just trying to project your silly opinion on others and hope to bend reality to your will by getting prophetic about the future...or you're trolling...probably both.
  • BrokenCrayons - Wednesday, February 22, 2017 - link

    I see you're back to once again try to decry the site. It's funny if things are so bad that you keep coming back to read articles, but that's a special sort of problem, I suppose. Anyway, since you're having trouble with reading comprehension, most of the comments swirl around the drive's poor performance relative to other products. Just because someone doesn't leap to their feet to applaud a device that landed at the bottom of a chart doesn't mean the reviewer, the article, or the company publishing it is doing something wrong by reporting their test results. Just because people are generally in agreement with the results via their comments certainly doesn't indicate there's a problem with the review. You're just trying to project your silly opinion on others and hope to bend reality to your will by getting prophetic about the future...or you're trolling...probably both.
  • watzupken - Saturday, February 25, 2017 - link

    So far, the Silicon Motion and Phison's controllers always seem to perform poorly with consistency. They don't seem like they are trying to fix it with each iteration of their controller.

Log in

Don't have an account? Sign up now