Intel DC P4510 2TB & 8TB Enterprise PCIe NVMe SSD Review (Page 5)

8K Random Write/Read

We pre-condition the drive for 16,000 seconds, or 4.44 hours, receiving performance data every second. We plot this data to observe the test subject's descent into steady-state. We plot both IOPS and Latency. We plot IOPS (represented by blue scatter) in thousands and Latency (represented by orange scatter) in milliseconds.

We observe steady-state for the 2TB model is achieved at 4,000 seconds of preconditioning. Average steady-state write performance at QD256 is approximately 44K IOPS. We observe steady-state for the 8TB model is achieved at 11,000 seconds of preconditioning. Average steady-state write performance at QD256 is approximately 77K IOPS.

Here again, we see the DC P4510 delivering double the random write performance of all the non-Intel contenders in our test pool at QD1. Both models lead the competition at QD1-QD2. Micron's 9100 MAX is a juggernaut at higher queue depths, but it could be argued that QD1-2 is a more important performance metric.

This means that depending on the workload, the DC P4510 likely delivers better write-heavy mixed workload performance at operating region queue depths than the write centric 9100 MAX. In fact, that's exactly what our mixed workload testing shows.

Just as we saw with our 4K testing, none of the competing SSDs in our test pool can match the DC P4510 until well after the majority of the typical operating region (QD1-9) has been exceeded. This means that even if the competing SSDs in our test pool exceed the performance of the DCP4510 at queue depths above 9, it doesn't really matter - the DC P4510 still wins.

Conclusion (TL;DR): Larger file size, same result. Intel's DC P4510 easily outperforms the competing SSDs in our test pool within typical operating region.