Random Read/Write Speed

The four corners of SSD performance are as follows: random read, random write, sequential read and sequential write speed. Random accesses are generally small in size, while sequential accesses tend to be larger and thus we have the four Iometer tests we use in all of our reviews. Our first test writes 4KB in a completely random pattern over an 8GB space of the drive to simulate the sort of random access that you'd see on an OS drive (even this is more stressful than a normal desktop user would see).

We perform three concurrent IOs and run the test for 3 minutes. The results reported are in average MB/s over the entire time. We use both standard pseudo randomly generated data for each write as well as fully random data to show you both the maximum and minimum performance offered by SandForce based drives in these tests. The average performance of SF drives will likely be somewhere in between the two values for each drive you see in the graphs. For an understanding of why this matters, read our original SandForce article.

Desktop Iometer - 4KB Random Read (4K Aligned)

Random read performance is excellent. The M3/Pro already had great random read performance to begin with but M5 Pro takes that one step further. We are looking at figures similar to Vertex 4 here.

Desktop Iometer - 4KB Random Write (4K Aligned) - 8GB LBA Space

At queue depth of 3, the M5 Pro is noticeably faster than its predecessors. The difference is 30MB/s (~18% on average), although Plextor still cannot match the performance of SandForce or Indilinx Everest 2.

Desktop Iometer - 4KB Random Write (8GB LBA Space QD=32)

Increasing the queue depth to 32 yields great results. Performance has significantly improved since the M3/Pro. The M5 Pro is on par with SandForce and Everest 2 based SSDs here.

Sequential Read/Write Speed

To measure sequential performance we ran a one minute long 128KB sequential test over the entire span of the drive at a queue depth of 1. The results reported are in average MB/s over the entire test length.

Desktop Iometer - 128KB Sequential Read (4K Aligned)

Sequential read performance remains unchanged from M3 Pro.

Desktop Iometer - 128KB Sequential Write (4K Aligned)

Sequential write is up when compared to M3 Pro, but for some reason, the M3 is still even faster than the M5 Pro. 

Inside the M5 Pro and Test Setup AS-SSD Incompressible Sequential Performance
Comments Locked

37 Comments

View All Comments

  • poccsx - Friday, August 31, 2012 - link

    Our it shop has installed about 50 Agility 3s. 2 were DOA and the rest have been fine. Agree with you on the PITA firmware update process.
  • stalker27 - Tuesday, September 4, 2012 - link

    I was wondering if Plex kept their reputation, where I'm from they kinds faded away from the market.
  • ksherman - Friday, August 31, 2012 - link

    With the M5S only seeming to be marginally slower than the Pro series, but a good bit cheaper, what would I be missing if I opted for an M5S drive?
  • DukeN - Friday, August 31, 2012 - link

    The e-Penis factor.
  • stalker27 - Tuesday, September 4, 2012 - link

    Is it vital?
  • Zak - Friday, August 31, 2012 - link

    5 year warranty versus 3 year?
  • poccsx - Friday, August 31, 2012 - link

    I don't know about you guys but I don't have anything in my computer that's older than 3 years.
  • Beenthere - Friday, August 31, 2012 - link

    The Pro version should be the only version Plextor releases IMO and they should be priced as the base model -which they are. The non-Pro versions are basically models without proper firmware to facilitate best performance. The base models are just under-achievers used to justify higher prices for the Pro models. Not cool IMO.

    BTW the minute theoretical gains do not make any substantial gains in typical operation so just buy whatever is on sale if you're ready to jump in a be an unpaid SSD beta tester as this is still very "immature tech" as Anand has stated.
  • bji - Friday, August 31, 2012 - link

    SSDs are not immature tech. Two years ago they were. I personally have not bought a mechanical hard drive in 2+ years and I don't ever intend to do so again.
  • sheh - Friday, August 31, 2012 - link

    You should beware of longer term data retention.

Log in

Don't have an account? Sign up now