Random Performance – Iometer (IOPS/latency), YAPT (random)

We are trying something different here. Folks tend to not like to click through pages and pages of benchmarks, so I'm going to weed out those that show little to no delta across different units (PCMark). I'm also going to group results performance trait tested. Here are the random access results:

Iometer:

Iometer is an I/O subsystem measurement and characterization tool for single and clustered systems. It was originally developed by the Intel Corporation and announced at the Intel Developers Forum (IDF) on February 17, 1998 – since then it got wide spread within the industry. Intel later discontinued work on Iometer and passed it onto the Open Source Development Lab (OSDL). In November 2001, code was dropped on SourceForge.net. Since the relaunch in February 2003, the project is driven by an international group of individuals who are continuesly improving, porting and extend the product.

Iometer – IOPS

iometer-ios-file.png

iometer-ios-web.png

iometer-ios-database.png

iometer-ios-workstation.png

Now we see where the Phoenix Blade truly surpasses the RevoDrive 350. There is a definite gain to be seen in the SBC technology at use here, as it can attain higher ultimate IOPS than OCZ's VCA 2.0 implementation.

Despite this gain over the competition, both solutions are fighting a losing battle in random IO performance when compared to just one modern SATA SSD at queue depths up to 8. Most desktop usage sits in this lower part of the chart, and the less latent SATA controllers are able to reach maximum speeds in this window, and before the Phoenix Blade and the RevoDrive even have a chance to really ramp up to full speed.

Some might believe the inclusion of a P3700 is a bit unfair here, but I disagree. The P3700 uses an 18 channel controller, but the G.Skill and OCZ solutions are using four 4-channel SandForce controllers (16 channels total). The reason the results are so different is mostly due to the significantly higher IO latency of SandForce, with the P3700 seeing added per-IO latency reductions thanks to its use of NVMe, which is significantly more efficient than SCSI or SATA.

Iometer – Average Transaction Time

For SSD reviews, HDD results are removed here as they throw the scale too far to tell any meaningful difference in the results. Queue depth has been reduced to 8 to further clarify the results (especially as typical consumer workloads rarely exceed QD=8). Some notes for interpreting results:

  • Times measured at QD=1 can double as a value of seek time (in HDD terms, that is).
  • A 'flatter' line means that drive will scale better and ramp up its IOPS when hit with multiple requests simultaneously, especially if that line falls lower than competing units.

iometer-trans-file.png

iometer-trans-web.png

iometer-trans-database.png

iometer-trans-workstation.png

The higher latency of the SandForce controller equipped units over other modern SATA SSDs is painfully clear here as well. The SATA units lose steam and become more latent after QD=8 mostly because they are saturating their respective busses by that point.

YAPT (random)

YAPT (yet another performance test) is a benchmark recommended by a pair of drive manufacturers and was incredibly difficult to locate as it hasn't been updated or used in quite some time.  That doesn't make it irrelevant by any means though, as the benchmark is quite useful.  It creates a test file of about 100 MB in size and runs both random and sequential read and write tests with it while changing the data I/O size in the process.  The misaligned nature of this test exposes the read-modify-write performance of SSDs and Advanced Format HDDs.

yapt-randreads.png

yapt-randwrites.png

The Phoenix Blade and RevoDrive 350 fare better as their controllers are not as sensitive to alignment. While this is labeled as a random test in YAPT, its block sizes are large enough to be considered borderline sequential for faster SSDs, which allows the top two units to blow past all of the SATA devices. The P3700 chokes hard on writes not aligned to 4k boundaries, and that definitely shows here.

« PreviousNext »