SSD Performance File Copy Tests

SSD Performance

Right, lets start off with our real-world file copy tests. Currently, certain controllers benefit from compressed files, while others don't. Certain storage units hate small files, others work well with them. So it only makes sense to do some manual tests on that. Any storage unit's nightmare, whether that is an HDD or SSD, is storing really small files as fast as possible.

File Copy

We start with a very simple file copy test. A 2.5GB uncompressed 'dummy' file is simply copied to and from the drive's root to the local PC, and then back again. This was done 3 times in order to garner a sense of consistency, and the drive was given minimal (at most, 10 seconds) time to rest in between those runs. This is relatively indicative of the work this kind of product might well be put through, i.e. single large file transfers between it and the PC it is plugged into. An alternative use for them is working off of the drive itself, e.g. editing and/or rendering photos or video with the file itself stored on the disk, and not the storage of the local PC.

Copying 'to' the drive from the local computer saw very consistent performance across all three runs, with the drive hitting anything between 320-340MB/s. That is, to my mind, very respectable, achieving almost the same speeds as the previously reviewed Seagate drive.

Copying 'from' the drive was significantly quicker, and noticeably so. Performance started out in the 440-450MB/s range and then - completely inexplicably - shot up to nearly 900MB/s..? I was absolutely flummoxed by this behaviour, and I cannot figure it out. This behaviour was consistently repeatable across a few different file types, as well as using different connections on my PC and both supplied types of cables. Absolutely bizarre. Someone more learned than myself could probably figure it out, I am sure. That aside, we are seeing very similar performance numbers to what we got out of the Seagate 'Fast' SSD we reviewed last year. I am not even remotely surprised by this, as I suspect both are probably using the same type of memory and - probably - controller. 

Next, we will move onto a favorite benchmark of the industry. ATTO.