Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

Looking at the benchmarks of two SSDs on today's market, one by Samsung and one by OCZ, I noticed something interesting between them. Samsung had faster reads than writes. While the Vector had roughly equal write/read throughput.

          Read      Write
Samsung  515 MB/s  491 MB/s
OCZ      497 MB/s  467 MB/s

Benchmark used is Blackmagicdesign's Disk Speed Test. I have noticed similar results with other SSDs.

Why is this the case? Is the read speed of a SSD equal to the write speed, or is it faster?

share|improve this question

closed as not a real question by Breakthrough, Diogo, 8088, Simon Sheehan, Louis Dec 17 '12 at 21:24

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form. For help clarifying this question so that it can be reopened, visit the help center.If this question can be reworded to fit the rules in the help center, please edit the question.

    
For overall performance is good to measure speed with different block sizes look for example at atto http://uk.hardware.info here are both of them with other models.. –  week Dec 17 '12 at 13:12
    
Um, if anything they Samsung's speeds are closer to being equal. –  Louis Dec 17 '12 at 21:23

3 Answers 3

When dealing with solid state drives (i.e. flash memory), the process of actually writing/erasing the data in a flash cell consumes more time than simply reading what occupies the cell. This is done by electron tunnel injection/release, over time causing a breakdown of the insulating oxide layer (reducing the lifespan of the drive, which is why we use TRIM).

However, over time, reading will slowly affect the voltages in the cell of interest itself, as well as the adjacent cells. Depending on how the SSD implements this, after some time, the adjacent cells will have to be rewritten just after the current read is finished, which can also affect performance. Unlike the constantly lengthier time it takes to erase/write a block, however, this effect is dependent on the manufacturer's implementation.

share|improve this answer

With the 840 that is expected behaviour - most drives have slightly faster read than write speeds - anantech has benchmarks that pretty much match yours, and i recall that with some other versions - such as the 'regular' TLC based 840, the firmware is designed to optimise read speeds even where the nand type is slower.

What you're seeing is by design since most people will notice write speeds more than read speeds.

share|improve this answer

All these numbers are more marketing stuff, than real information. These numbers are achieved only with very special sized files, and only on limited amount. In real life these numbers much lower(if this word exists). So few megabytes of difference wont play big role

share|improve this answer

Not the answer you're looking for? Browse other questions tagged or ask your own question.