HP FX900 1 TB NVMe Review

Memory (DDR4/DDR5) and Storage (SSD/NVMe) 368 Page 9 of 17 Published by

teaser

SSD Performance Linear Read and Write performance analysis

SSD Performance Linear Read and Write performance analysis

 In this chapter of the review, we will fill the SSD with data and observe what happens to the write performance due to doing so. So, for example, if the SSD has a capacity of 2TB, we'll read and then write continually. So we can see how the SSD would react simultaneously when subjected to large amounts of reading and writing activities.

As a solution to the sluggish read/write performance of TLC SSDs, manufacturers have added an SLC cache, sometimes known as a pSLC cache, into their products (pseudo). This test can examine what happens when the cache fills up and then write speed falls back to the lowest possible value. The amount of space available is determined by the capacity of the SSD. If there is still some free SLC cache space available, the SSD's read/write performance is comparable to that of SLC NAND, as long as the area has not been utilized. When the SLC cache is completely depleted, the firmware (FW) initiates background garbage collection to clear off the remaining space. Additionally, while garbage collection is taking place, data is still being sent to the drive, adversely affecting its overall performance.

 

Page9_1


 The read performance remains constant across the entire storage capacity; we average out at ~5800 MB/sec on average throughout the whole linear reading process (which is more than the theoretical output of 5000 MB/s!)

 

Page9_2


Linear write testing then; When it comes to writing performance on any NAND storage device, the infamous Achilles heel comes into play - if the pSLC buffer runs out of space, and once that happens, you're back to bare TLC write performance. We peak at ~4600 MB/sec and after ~30% (300GB) percent written on the SSD, writes drop, and average around 1200 MB/s (which is still over two times more than the best SATA SSDs). 

Share this content
Twitter Facebook Reddit WhatsApp Email Print