Samsung SSD 840 EVO mSATA (120GB, 250GB, 500GB & 1TB) Reviewby Kristian Vättö on January 9, 2014 1:35 PM EST
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). I perform three concurrent IOs and run the test for 3 minutes. The results reported are in average MB/s over the entire time. We use both standard pseudo randomly generated data for each write as well as fully random data to show you both the maximum and minimum performance offered by SandForce based drives in these tests. The average performance of SF drives will likely be somewhere in between the two values for each drive you see in the graphs. For an understanding of why this matters, read our original SandForce article.
As expected, random IO performance is similar to the original EVO. There is some slight variation of course but nothing that stands out.
Sequential Read/Write Speed
To measure sequential performance I ran 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.
For some reason, the 1TB EVO mSATA is a few percent faster in 128KB sequential read test but falls short in the sequential write test. It's possible that the 16-die NAND has some effect on performance, which would explain the difference, but we're still dealing with rather small differences.
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.
Post Your CommentPlease log in or sign up to comment.
View All Comments
GreenThumb - Monday, February 10, 2014 - linkHow would I know if a new laptop w/ Haswell processor could accept one of these drives?
mdevlin - Friday, February 14, 2014 - linkOne important point that was not captured in the review is thermal limiting. Our experience is that when doing large transfers (200GB+) the 1TB msata 840EVO starts to slow down and will sometimes run at 100MB/s or slower whereas we have not had this problem with the 2.5" 1TB 840 EVO. Doing a better job of cooling (even just fanning the air above a drive under test) improves mSATA performance. Not a big deal if you do not do large contiguous writes. We have never had any problem with Read performance.
tomb18 - Monday, March 10, 2014 - linkHas anyone actually bought one of these 1TB drives?
I have bought 2, from two different locations about 6 weeks apart. If you copy a large quantity of data (over 400gB) from another disk to one of these drives, they will fail during the write. They disappear completely, are not visible at all unless you do a reboot. Thje 500GB works fine.
I have verified this now on two different SATA3 controllers and on three different Windows 7 computers. Both 1TB drives fail on write.
RMA is in process with Samsung. I was told it sounds like a controller issue on the SSD.
mdevlin - Friday, March 14, 2014 - linkWe have 6 of the 1TB mSATA drives and have no problem with large writes (up to 800GB) except the thermal issues mentioned above. With forced air cooling the performance is sustained throughout the transfer. without cooling speeds drop to less that 100MB/sec and it will actually pause and then resume (but not fail). we have had no failures after many full disk writes.
mdevlin - Tuesday, April 8, 2014 - linkIt seems that Samsung has updated the firmware on the 1TB mSATA 840EVO. Now when the mSATA drive overheats it throttles by alternately halting for a second, then running full bore for a second. This is less disconcerting than the previous version which would slow down for long periods or even stop for 10-20 seconds.