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
Interesting results here. In the web server test, which is mostly reads, the Neutron XT sits among the best SSDs out there – at queue depths up to 4. After that, it essentially flattens out. It does so prior to reaching its SATA throughput limit, indicating the controller is an early limit here. The other charts show a similar leveling off after QD=4, resulting in even lower performance in the mixed R/W workloads.
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.
Although we end the latency charts end at QD=8, we can see the XT creeping up with higher latencies than the others in many tests.
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.
Impressive results seen in YAPT here. Although this is a 'misaligned test', the XT plows through, turning in the highest results. What YAPT considers as random appears more sequential to an SSD, and the Neutron XT does very well with bulk sequential transfers.
Thanks for the excellent
Thanks for the excellent review. It is probably me, but I find it hard to read confidently the io charts. Maybe you can make it so we can click the device on the right to turn on and off its curve. Then we could put one or more of them on the chart at a time.
“My inclination is to say that this is a great drive for the money, but without the actual pricing, the jury will remain out on that verdict until the drive actually launches,” you stated at the end. How can you make a qualitative statement about something you have no idea about? That is very misleading no matter the qualifications.
Allyn, do you have any
Allyn, do you have any thoughts on the suggestion in the first paragraph about the charts?
My ‘inclination’ is based on
My 'inclination' is based on the fact that it's a Neutron series SSD, and those have been cost-effective in the past.
On the charts, we're looking into better ways to display more data (dynamically even), but for now we're doing static images. I'll see if I can get a bit more distinction among the charted lines.
Maybe you could make each
Maybe you could make each name a js button that displays or hides the associated curve on the chart, or displays or hides different variations of the chart, but the latter is not practical.
At first I thought that the
At first I thought that the S3110 choking on certain file sizes was bizarre. However, after seeing the benchmarks around the web for the initial prototype the same dips were there. There’s been some tuning since then obviously.
I like the fact that they took at crack at accelerating compressed samples while maintaining competent speeds with incompressible files. Sandforce has had trouble in that area.
This is just another SATA 3 drive, and an oddly inconsistent one at that. It just comes out to be seemingly generic to me.
Yeah, their ‘acceleration’
Yeah, their 'acceleration' seems inconsistent, and is only effective on very compressible data (i.e. repeating patterns). This doesn't come up very often on an active system, as the majority of the repeating patterns would be 0's from TRIMmed areas, but those wouldn't be tracked in flash regardless.
ive had nothing but bad luck
ive had nothing but bad luck with Corsair Neutron drives in the past. Here’s hoping these new ones are more reliable.
I just don’t know how you can
I just don’t know how you can put up an SSD review without knowing pricing?
Well, the other option was to
Well, the other option was to not publish a review, but then we would not be covering an SSD that other review sites have covered…
I wonder how relative the
I wonder how relative the performance of this controller is to the Phison controller in the “Amazon Exclusive” Patriot Torch SSD that IS currently available for purchase…