Article index
Author
IOMeter test results
I/O Performance
There is little point of having an SSD drive that has blazing sustained reading and writing speeds, if the drive can’t handle reading and writing of small random files. If you intend to use your new SSD drive to store and run your operating system, then the drive must be able to cope with the many small random files that Windows will write to the drive continually. So we feel it is very important to test how many of these random files that a drive can handle in one second. We believe that anything over 1000 I/O’s per second would be enough for most users running a consumer grade mainstream PC, and should provide a smooth running system. But obviously, the more I/O’s that a drive can handle, the faster the drive will feel and leave more headroom for those huge multitasking sessions that users sometimes engage in.
The things that we will look at are the total I/O per second and total MB/s.
Partition alignment and sector boundaries
Windows 7 will automatically align a partition during partition creation, Windows XP won’t. It is imperative that the SSD’s partition is aligned. Windows XP is also restricted to sector boundaries, while Windows 7 will use 4k boundaries if it can.
IOMeter allows us to set the sector boundaries for conducting the tests, and we have therefore set the sector boundaries at 4K, which means the IOMeter tests are valid for Windows 7 and Windows Vista users. XP users will not be able to obtain such results.
We will provide a screenshot of the tests on the review drive for those of you who like to see the actual test result. All the comparison drive results are represented in the form of graphs.
If any of you would like to see a screenshot from any IOMeter test on a particular drive, please feel free to request one, and we’ll post the screenshot in the forum thread.
IOMeter 4K random (outstanding I/Os = 4) write test
IOMeter 4K random (outstanding I/Os = 4) write test.
Our first test involves creating continual 4KB random files on the target drive with IOMeter. We use a 4KB file size, as it is believed that Windows will create and modify many of this size of file constantly in the background during a typical Windows session.

Transcend 32GB SSD model

Transcend 256GB SSD model

IOMeter – IOPS

IOMeter – MB/s
When it comes to dealing with 4KB writes, the Transcend 256GB model falls well behind most of the Marvell and Sandforce based SSDs, with the 256GB model just keeping up with the Indilinx based drives. Also, while the Transcend 256GB model managed to give some impressive sequential results, this test clearly shows that when it comes to random 4KB write handling, it simply can’t compete.
On the other hand, when compared with the two traditional hard disks, both the 32GB and 256GB models clearly blew all the HDDs away with their performance. The 32GB model falls well behind the 256GB SSD, but still manages to beat the previous 32GB model by double.
One thing we did notice while carrying out this test is that as the test progressed, the IOPS and transfer rates fell off as with the previous Transcend SSDs, with a large drop after 45 seconds for both models. We tried testing at different queue depths to see what impact this made, but the outcome was much the same each time. To give an idea, we recorded the IOMeter test with a 5 second update frequency, running the 4K Random write test for 5 minutes. In the following graph, the red line is for a queue depth of 1 and the blue line is for a queue depth of 4:

Transcend 32GB SSD model – IOPS

Transcend 256GB SSD model – IOPS
What this graph means is that if the drive is subject to a lot of continuous random write activity, such as heavy swap file usage or a large software package installation, the write performance will seriously degrade. The performance recovers after the drive is left idle, for example, we ran the second queue depth test after leaving each SSD idle for 30 minutes. We will take a look into the write performance of small files later in this review.
IOMeter 4K random (outstanding I/Os = 4) read test
If there are many 4k files created, then that must also mean that many 4k files need to be read. This test measures 4k reading performance.

Transcend 32GB SSD model

Transcend 256GB SSD model

IOMeter – IOPS

IOMeter – MB/s
As we noticed in our earlier benchmarks involving 4KB reads, both Transcend models struggle when it comes to reading blocks of this size. Even still, no traditional spinning hard disk today can rival these figures. For example, both SSD models give results well over 10 times better than any hard disk we tested here. Interestingly, the 32GB model gives significantly better 4K reading than the 256GB model, something we noticed earlier in the AS SSD and CrystalDiskMark tests.
IOMeter 512KB (outstanding I/Os = 4) write test
Sequential writing performance is also very important.

Transcend 32GB SSD model

Transcend 256GB SSD model

IOMeter – IOPS

IOMeter – MB/s
Like our earlier sequential write tests, the Transcend 256GB has a very impressive sequential write performance, only beaten by the OCZ Vertex 2 drive in this test. The 32GB model on the other hand is the slowest drive in this test, not even managing to beat the 2.5” HDD here.
IOMeter 512KB (outstanding I/Os = 4) read test
This test measures 512KB sequential reading performance.

Transcend 32GB SSD model

Transcend 256GB SSD model

IOMeter – IOPS

IOMeter – MB/s
Again, an impressive read performance from the 256GB SSD, beating most of the drives. Like the write test, the 32GB SSD is the slowest SSD in this test, only beating all but one of the spinning hard disks.
IOMeter Workstation simulation (outstanding I/Os = 64)
When running applications you will find that there is a mixture of small random files, and larger sequential files being created and read. Not only that, it isn’t just one file at a time. In this test we measure a simulated workstation pattern, with a queue depth of 64 (threaded).

Transcend 32GB SSD model

Transcend 256GB SSD model

IOMeter – IOPS

IOMeter – MBs
It’s worth noting that the IOMeter workstation test is more severe than probably any amount of activity one would carry out in the real world, maybe with the exception of a busy database or file server. As we can see here, both SSDs really struggle in this test, but again like our random 4KB write test, both still perform several times better than all the HDDs tested here.
Summary
Unlike the earlier benchmarks, these IOMeter tests clearly show some of the strengths and limitations each SSD has. While the 256GB model did extremely well in the sequential tests, it fell way behind most of the SSDs in the other tests.
Again as we’ve mentioned many times before, all the tests we have carried out up until this page are synthetic tests and can only be used as a rough guide to how these SSDs will perform in real life use. No amount of testing on a car will reveal how it will feel and handle on the road with the driver behind its wheel and the same goes with an SSD.
So let’s head on to the next page where we check out some real world testing, starting with a fresh clean Windows 7 installation…
Latest Comments
Hot Threads
- DoMiN8ToR, last post: alan1476 20
- StormJumper, last post: jsoulesone 13822
- coolcolors, last post: alan1476 9
- 2601, last post: jadburner 50
- Clueless in Seattle, last post: AllanDeGroot 15

