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
The Silicon Motion controller drives are outmaneuvered by the Samsung 850 and Intel 730 in some tests, but overall perform very well. They do fall slightly behind the M550 and MX100, but are extremely close.
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.
No surprises in transaction times for the BX100.
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.
While the MX100 had a hard time in our YAPT random read test, the BX100 does not suffer from this same issue.
This test has no regard for 4k alignemnt, and it brings many SSDs to their knees rather quickly. The BX100's handled it as well as could be expected given the flash architecture they employ, and come in mid-pack. As an aside, the Intel SSD 730 *really* does not like writes not aligned at 4k boundaries.
It’s a sad day when $0.36/GB
It’s a sad day when $0.36/GB (on a budget drive, even) is considered “impressively low”. Six years ago, those prices were already completely unacceptable. HDD manufacturers really need to step it up, because SSD manufacturers certainly aren’t doing it.
Fact the C300 launch price
Fact the C300 launch price was $800 for 256 gig ssd 5 years ago that’s more than $3.7 per gig it quickly fell to $650 around this time 5 years ago via sales though the price drop is more than 10 times lower per gig for the bx100, and just like the conclusion states this will likely cause a ssd price war. But I believe we will start seeing even more dense ssd for them to make up for the revenue loss.
It’s a sad day when $0.36/GB
$0.36/GB _is_ impressively low for an SSD. NAND flash is just a more expensive technology than magnetic spinning disks right now. If you don’t like it, too bad.
They’re trying their best, and doing a good job I might add. I don’t see you helping. Anyone can be a critic…
Allyn, I am always curious
Allyn, I am always curious to know if tools like the Crucial Storage Executive function the same when multiple SSDs are members of a RAID array wired to a third-party controller.
Popular websites like http://www.pcper.com could help end users by urging third-party RAID controller vendors to support TRIM at a minimum. Intel’s RST has supported TRIM for some time now.
Perhaps the industry in general should confront what needs to happen to standardize SSD maintenance tasks, so as to inter-operate across platforms, motherboards, chipsets and add-on controllers.
Plug-and-Play for SSDs!
Thanks again for another good review.
MRFS (not just dreamin’ this time 🙂
TRIM through a RAID is still
TRIM through a RAID is still a tricky thing. It took Intel some time even despite my repeated urging to do so. It's also does not work on parity arrays (RAID-5/6).
Accessing SMART and other management through third party RAID is tricky. Areca actually does it, but you must use their API to request SMART data from their cards.
Wake me up when a “512GB of
Wake me up when a “512GB of quality SSD-space for ~92$” would become a mainstream thing. That’s how much I’ve paid for my 512GB MX100 when it just came out, so…
Isn’t the MX100 a better SSD?
Isn’t the MX100 a better SSD? Reason I ask because there is only a $10 to $20 difference depending on where you purchase.
BX100 gets faster write
BX100 gets faster write speeds at smaller capcities when compared to the MX100. Once you hit 500+GB everything evens out. The SM controller in the BX100 is also a bit faster than the MX100 in the seqential reads.
Long story short, it's a newer generation controller driving faster flash than the MX100, so the BX100 is better, which is why Micron is replacing the MX100 with the BX100 and introducing an MX200 (which works differently).