A couple of months ago we reviewed JMicron's JMF667H reference design SSDs, which did relatively well in our tests especially when taking JMicron's previous SSD controllers into account. As always, reference designs are only meant for evaluation and do not make their way into retail, so today we are taking a look at Transcend's SSD340.

The SSD340 is based on the same JMF667H controller, although the firmware is an older version than what we tested in the reference design SSDs. Transcend told us that they currently have no plans to update the firmware to the newer version, which I am guessing is due to using customized firmware. It is actually quite rare that OEMs use the stock firmware as typically the OEMs request some changes and customizations, making it slower and more difficult to upgrade the firmware. I would not even be surprised if some OEMs did not upgrade the firmware to ensure product differentiation because it is obvious that OEMs do not want their low end drives to jeopardize the sales of higher cost (and profit) drives. 

Update 8/18: Apparently Transcend has released a new firmware for the SSD340. I will try to find some time to test the new firmware and will provide an update if anything changes. 

The SSD340 is available in capacities ranging from 32GB to 256GB. The JMF667H actually has a capacity limit of 256GB, which also limits the SSD340 to just 256GB. We should see JMF670H make its entry later this year with support for 512GB, but until then 256GB is the highest you can go with a JMicron controller.

Like nearly all client SSDs, the SSD340 has a three-year warranty with an endurance limitation. Quite surprisingly the endurance scales with capacity, although the scaling is not exactly linear. Nowadays most client SSDs only have a single rating for all capacities, so it is pleasant to see a scalable endurance for a change. I am pretty sure many OEMs just artificially lower the spec to make sure that enterprises customers do not choose the lower profit consumer drives, as the difference between high-end client and entry-level enterprise drives is quite indeterminate (i.e. entry-level enterprise drives are usually based on client platforms).

Transcend's data sheet for the SSD340 states that the drive supports DevSleep, although there are not any actual power figures listed aside from normal idle, which are way too high for DevSleep. Since the controller supports DevSleep, I have to wonder why Transcend has not published any power figures; we'll check power later on to verify whether the SSD340 actually supports DevSleep.

Transcend uses Micron's 128Gbit 20nm NAND in the SSD340. Our 256GB review sample had a total of sixteen NAND packages (eight on each side of the PCB), meaning that each package is a single-die package with one 128Gbit (16GB) die. There is also Samsung's 256MB DDR3-1600 DRAM chip working as a cache. 

Test System

Thanks to G.Skill for the RipjawsX 32GB DDR3 DRAM kit

Performance Consistency

Performance consistency tells us a lot about the architecture of these SSDs and how they handle internal defragmentation. The reason we do not have consistent IO latency with SSDs is because inevitably all controllers have to do some amount of defragmentation or garbage collection in order to continue operating at high speeds. When and how an SSD decides to run its defrag or cleanup routines directly impacts the user experience as inconsistent performance results in application slowdowns.

To test IO consistency, we fill a secure erased SSD with sequential data to ensure that all user accessible LBAs have data associated with them. Next we kick off a 4KB random write workload across all LBAs at a queue depth of 32 using incompressible data. The test is run for just over half an hour and we record instantaneous IOPS every second.

We are also testing drives with added over-provisioning by limiting the LBA range. This gives us a look into the drive’s behavior with varying levels of empty space, which is frankly a more realistic approach for client workloads.

Each of the three graphs has its own purpose. The first one is of the whole duration of the test in log scale. The second and third one zoom into the beginning of steady-state operation (t=1400s) but on different scales: the second one uses log scale for easy comparison whereas the third one uses linear scale for better visualization of differences between drives. Click the buttons below each graph to switch the source data.

For more detailed description of the test and why performance consistency matters, read our original Intel SSD DC S3700 article.

	Transcend SSD340	JMicron JMF667H (Toshiba NAND)	JMicron JMF667H (IMFT NAND)	Samsung SSD 840 EVO mSATA	Crucial MX100
Default
25% OP

Right from the start, things do not look too promising. Compared to the reference design with the same IMFT NAND, the IO consistency is considerably lower. The reference design manages around 1,000 IOPS minimum, whereas in the SSD340 the minimum performance is around 300 IOPS. Increasing the over-provisioning helps by a bit but the consistency is still poor compared to the other value drives (like the MX100). The older firmware definitely isn't doing Transcend any favors here – quite the opposite in fact.

	Transcend SSD340	JMicron JMF667H (Toshiba NAND)	JMicron JMF667H (IMFT NAND)	Samsung SSD 840 EVO mSATA	Crucial MX100
Default
25% OP

	Transcend SSD340	JMicron JMF667H (Toshiba NAND)	JMicron JMF667H (IMFT NAND)	Samsung SSD 840 EVO mSATA	Crucial MX100
Default
25% OP

TRIM Validation

To test TRIM, I first filled all user-accessible LBAs with sequential data and continued with torturing the drive with 4KB random writes (100% LBA, QD=32) for 30 minutes. After the torture I TRIM'ed the drive (quick format in Windows 7/8) and ran HD Tach to make sure TRIM is functional.

And it is. 

AnandTech Storage Bench 2013

Our Storage Bench 2013 focuses on worst-case multitasking and IO consistency. Similar to our earlier Storage Benches, the test is still application trace based – we record all IO requests made to a test system and play them back on the drive we are testing and run statistical analysis on the drive's responses. There are 49.8 million IO operations in total with 1583.0GB of reads and 875.6GB of writes. I'm not including the full description of the test for better readability, so make sure to read our Storage Bench 2013 introduction for the full details.

We are reporting two primary metrics with the Destroyer: average data rate in MB/s and average service time in microseconds. The former gives you an idea of the throughput of the drive during the time that it was running the test workload. This can be a very good indication of overall performance. What average data rate doesn't do a good job of is taking into account response time of very bursty (read: high queue depth) IO. By reporting average service time we heavily weigh latency for queued IOs. You'll note that this is a metric we have been reporting in our enterprise benchmarks for a while now. With the client tests maturing, the time was right for a little convergence.

Ouch, this is bad. The SSD340 is the slowest SSD in our 2013 Storage Bench and the difference is quite substantial. Given the results on the previous page along with the results here, it is clear that the SSD340 is not suitable for IO heavy workloads that require consistency. It still might have something to offer in lighter workloads, but that's looking increasingly doubtful.

AnandTech Storage Bench 2011

Back in 2011 (which seems like so long ago now!), we introduced our AnandTech Storage Bench, a suite of benchmarks that took traces of real OS/application usage and played them back in a repeatable manner. The MOASB, officially called AnandTech Storage Bench 2011 – Heavy Workload, mainly focuses on peak IO performance and basic garbage collection routines. There is a lot of downloading and application installing that happens during the course of this test. Our thinking was that it's during application installs, file copies, downloading and multitasking with all of this that you can really notice performance differences between drives. The full description of the Heavy test can be found here, while the Light workload details are here.

Fortunately the performance is much better in our 2011 Storage Benches. It looks like the JMF667H does fairly well in terms of peak performance but as soon as you fill the drive and start playing with consistency, the performance will take a massive hit.

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.

Random performance is okay and similar to the other JMF667H SSDs. Queue depth scaling has never been JMicron's strength but then again typical client workloads do not usually go above queue depth of 5 anyway.

Sequential Read/Write Speed

To measure sequential performance we run a 1 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.

Sequential performance leaves a lot to be desired. Especially write performance is poor for a 256GB drive.

AS-SSD Incompressible Sequential Read/Write Performance

The AS-SSD sequential benchmark uses incompressible data for all of its transfers. The result is a pretty big reduction in sequential write speed on SandForce based controllers.

Performance vs. Transfer Size

ATTO is a useful tool for quickly benchmarking performance across various transfer sizes. You can get the complete data set in Bench. Read performance scaling is actually better than in the reference design JMF667H we tested, although the SSD340 is still outperformed by the 840 EVO and SP610. Write performance seems to be similar to the reference design and the maximum throughout is definitely a bit disappointing. 

Click for full size

Power Consumption

It looks like the SSD340 does not support slumber power states (HIPM+DIPM), or at least it is not very efficient in doing so. Load power consumption looks better, although the reference design is still more power efficient. Given the lackluster performance so far, this is just one more reason to look elsewhere.

Final Words

From a performance perspective, the SSD340 leaves a lot to be desired. It is the slowest SSD in our 2013 Storage Bench and the IO consistency is also quite horrible compared to the competition. I have to wonder why Transcend is not utilizing the newer firmware from JMicron because the reference design SSD with Intel's 128Gbit 20nm NAND is much faster than the SSD340 is. It is certainly possible that Transcend is using lower grade NAND to cut costs, which would explain the lower performance, but I find it hard to believe that the NAND alone would result in up to 35% decrease in performance. 

Ultimately it all boils down to price and that is where the SSD340 fails to set itself apart from the competition. The SSD340 is definitely one of the cheapest SSDs around but the competition can provide a much better feature set and performance at a similar price. For the price of the SSD340, you can get ADATA Premier SP610, SanDisk Ultra Plus or Crucial MX100 – all of which are better picks than the SSD340. The only advantage that the SSD340 has is the 64GB model that most manufacturers no longer offer, but I would strongly recommend spending $15-20 more to get twice the capacity and a better SSD (e.g. the MX100). 

All in all, the SSD340 is a rather unimpressive drive. At $50 for 128GB and $90 for 256GB, it might be a good option for buyers that have a very tight budget, but at the current prices the SSD340 just does not make any sense. You are much better off with the Crucial MX100 or ADATA Premier SP610 at the same price.