Filling up the SSD with data
For obvious reasons, when an SSD is tested, the drive is
always tested as a spare drive, and is generally always empty (no data on the
drive) during the synthetic benchmarks. There is no other way of having a level
playing field for all the SSDs under test. This of course changes during the
real world tests we conduct here at Myce.com.
Real users of course don’t buy an SSD for it to remain
empty, and how full the SSD will eventually become varies from one user to the
next. What I thought would be useful is to run tests on the SSDs with real data
on the drives, and at different levels regarding how full the drive is.
Myce Sustainable Performance Test
Over the last few months I have been studying countless
analyzer traces of real computing workloads, and also developing a test that
would accurately emulate and measure how performance is sustained over a period
of time. For obvious reasons, it is not possible to test an SSD review sample
over several months before publishing a review. The solution was to condense
this down to a manageable test, that doesn’t take too long to run.
I will make it clear right from the outset that this is not
a torture test. Bringing any SSD to its knees is not helpful in the least, as I
for one would not use any SSD that had slowed down to crawl, just to prove a
point. The Myce Sustainable Performance test, I believe is a tough, but
sensible test pattern to use for measuring how an SSD will be behave once it’s pushed
hard over a period of time.
The test pattern is "workstation" based, and
closely emulates a typical video or graphics workstation environment. The
results are measured using the same hardware I use for the Myce Reality Suite
tests, however, the test data and measuring system use a different method.
With the drive filled to 80% of its capacity, I already have
an SSD with a lot of data on it. Adding to the data that is already there, the
"Sustainable Performance" test data is added. This test data is
approximately 20GB is size, so once this is added the SSD is pretty full.
The test is then run for a period of 20 minutes. 60
performance measurements are taken for every minute of the test, and an average
performance figure is generated after each minute. At the end of the test I
have 20 performance measurements which are then used to generate the graph
The faster SSDs will obviously sustain more writes then the
slower SSDs. For the fastest SSD in this test, the test pattern generated 146GB
of writes, and 193GB of data was read from the SSD during the test.
When reading the graph, you should not pay too much
attention to which drive is the fastest, but instead look at the sustainable
performance curve of each SSD, as this is what this test is all about.
For the SSD that I am reviewing, I will also add a second
graph which looks at the result in more detail.
So let’s look at the results.
Detailed results for the
There is evidence of a slowdown in the performance of the Samsung
750 EVO SSD. It doesn’t slow down by a huge amount though, and in fact it is in
the region of 70 MB/s.
Sustainable writing performance
Whilst the ‘Myce Sustainable Performance Test’ is based on a
mixed reading and writing workload, I believe it to be a good representation of
how an SSD will perform under a heavy workload, and I am often asked how a
particular SSD holds up performance wise, when it is subjected to a heavy data
writing workload. The fact is all SSDs aimed at the consumer market will slow
down to a crawl if you absolutely murder them with constant ‘writes’. Murdering
a consumer based SSD with writes tells us only what we expect, and is therefore,
in my opinion, a totally pointless exercise.
Most of us consumers can write quite a lot of data in a
single session, but very few of us will write lots of data constantly. If you
do write lots of data continually, then an SSD aimed at consumers is not for
you, and you should seriously consider spending the extra money required to
purchase an enterprise class SSD.
I have given quite a lot of thought to how to conduct a
sustainable writing performance test that is both meaningful, and uses a real
world scenario, that you are likely to come across as a consumer. One should
always make a system backup. SSDs tend to fail without notice, and if you don’t
a have a system backup when an SSD fails, then I’m afraid you’re in for a long
haul getting your operating system installed and updated, and then
re-installing all your apps.
So for this test, I will measure the time taken for a
popular backup application to ‘restore’ a system backup image to the target
SSD. Using the restore feature will subject the target SSD only to ‘writes’,
and providing the ‘restore image’ is large enough, then it should provide a
good idea of how well the target SSD will perform when asked to write a large
amount of data in a single session.
For this test I will be using Macrium Reflect (free
edition) to first make a system image of my review PC. Having done this, the
image created is now set in stone and will be used for future reviews. The
system backup image was then copied to an OCZ REVODrive 350 PCIe SSD, and a backup
copy stored on an HDD for safe keeping. As the REVODrive 350 will be used to
read the system restore image, it has to be fast enough at reading data to
ensure that the reading drive is not slowing down the target writing drive. I
have complete confidence that the REVODrive 350 has more than enough reading
performance to ensure this test can remain constant for the foreseeable future.
The system image is 95.73GB on the drive, which is small
enough to fit onto a 120GB SSD, and yet large enough to give a fair
representation of sustainable writing performance on a larger capacity SSD.
All target SSDs in this test were secure erased before
running the test to make sure the test is fair. Since this is the first time I
have included this test in my review, I will show a screenshot of both the SSDs
used in this test. In future reviews, I will only show a screenshot of the
‘review’ SSD, and all other obtained results will displayed in a graph.
The results will be ‘time’ based. Or in other words, the total
amount of time taken in seconds to restore the ‘system backup image’ to the
Please note that in the screenshot, the OCZ Vector
(my current system drive) is shown as the source drive, as this was the drive
that was backed up. This was not the reading drive in the test. The reading
drive was an OCZ REVODrive 350.
Macrium Reflect system restore
system restore – Samsung 850 Pro 500GB
system restore – Samsung 750 EVO 500GB
Let’s look at the results in the graph below.
It isn’t a surprise to see that the 850 Pro is faster than
the 750 EVO.
Let’s head to the next page for an in depth analysis of
power consumption and efficiency………..