Intel's X25-V & Kingston's 30GB SSDNow V Series: Battle of the $125 SSDs
by Anand Lal Shimpi on March 19, 2010 12:00 AM EST- Posted in
- Storage
Random Read/Write Speed
This test reads/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). I perform three concurrent IOs and run the test for 3 minutes. The results reported are in average MB/s over the entire time.
I've had to run this test two different ways thanks to the way the newer controllers handle write alignment. Without a manually aligned partition, Windows XP executes writes on sector aligned boundaries while most modern OSes write with 4K alignment. Some controllers take this into account when mapping LBAs to page addresses, which generates additional overhead but makes for relatively similar performance regardless of OS/partition alignment. Other controllers skip the management overhead and just perform worse under Windows XP without partition alignment as file system writes are not automatically aligned with the SSD's internal pages.
First up is my traditional 4KB random write test, each write here is aligned to 512-byte sectors, similar to how Windows XP might write data to a drive:
Intel's drives have traditionally done very well in random write tests and the X25-V is no different. Here it performs like a much more expensive X25-M G2. The SSDNow V Series Boot Drive however is at the bottom of the charts in single digits here.
Running 4K aligned writes (similar to Windows 7 or OS X 10.5/6) doesn't improve the situation much in either case:
Random read performance is equally strong for Intel:
And equally weak for Kingston.
Facebook
Twitter
RSS
78 Comments
View All Comments
Taft12 - Saturday, March 20, 2010 - link
A few hundredths of a second - $1000 well spent!!!semo - Friday, March 19, 2010 - link
wow am i seeing <2mb/s sequential write speed on those drives? it must be an error i'm pretty sure that SD cards can do better than thatsamspqr - Friday, March 19, 2010 - link
there's a 4 on the other side of the axis: it's 41mb/sQChronoD - Friday, March 19, 2010 - link
That is annoying...Hey Anand (or TWIMC)
Isn't there a way for your pretty graph program to move the text to right of the bar if there isn't enough room for the label to fit? It's really annoying when you can't read half of the results because the super long product names are compressing the graph...
icrf - Friday, March 19, 2010 - link
Agreed, it is pretty annoying. What is used to generate these graphs? A third party tool or something in house? What language/what tool?Taft12 - Saturday, March 20, 2010 - link
I remember one of the other regular contributers mentioning that getting the graphs to look nice is not trivial, and I've had the same experience trying to accomplish the same.Of course, if the bar is so small you can't see the number, it doesn't really matter what the number is, the result is dismal.
Anand Lal Shimpi - Friday, March 19, 2010 - link
It's an automated system that unfortunately doesn't handle that case very well. We're launching the new site in a few weeks and we'll be updating the graph styles as well, so I'm going to try to get that fix in there :)Take care,
Anand
fless - Friday, March 19, 2010 - link
You can change the location of those labels in swiff chart pro. Right click on the number then select single label. You're too lazy.samspqr - Friday, March 19, 2010 - link
hisome time ago, 3dprofessor found that swapping a standard HDD for a WD velociraptor increased rendering performance by a sizeable 10%
http://www.3dprofessor.org/Reviews%20Folder%20Page...">http://www.3dprofessor.org/Reviews%20Folder%20Page...
in fact, I'm pretty sure in the core2 times they got much bigger improvements on a dual-socket board, but I can't find that review anymore
you also do rendering tests on your CPU reviews
could you test rendering performance of some standard 7200 drive, vs the velociraptor, vs a low-end SSD, vs a high end SSD? with your new gulftown, perhaps?
thanks
AnnonymousCoward - Saturday, March 20, 2010 - link
Anand stopped benchmarking SSDs in ways that pertain to real life long ago. http://tinyurl.com/ylflfao">http://tinyurl.com/ylflfao