HP SSD EX920 1TB M.2 Drive Review

By

ATTO & CrystalDiskMark

ATTO v3.05

ATTO is one of the oldest drive benchmarks still being used today and is still very relevant in the SSD world. ATTO measures transfers across a specific volume length. It measures raw transfer rates for both reads and writes and places the data into graphs that can be very easily interpreted. The test was run with the default runs of 0.5KB through 64MB transfer sizes with the total length being 256MB.

ATTO – HP SSD EX920 M.2 1TB:

Benchmark Results: ATTO showed the HP SSD EX920 M.2 1TGB PCIe NVMe drive reaching speeds of up to 3123 MB/s read and 1801 MB/s write in the standard overlapped I/O benchmark. This drive is rated at up to 3200 MB/s max sequential read and 1800 MB/s max sequential write, so we just missed the rated read speed.

Benchmark Results: Compared to some other SATA III and PCIe NVMe SSDs, the HP SSD EX920 M.2 1TB drive performs well and tops at over 3 GB/s at the higher block sizes and is in the top five for peak sequential write speeds.

Benchmark Results: The HP SSD EX920 M.2 1TB drive has pretty good sequential write speeds and is in the top 5 drives again for peak write speeds.

CrystalDiskMark 6.0.0 x64

CrystalDiskMark is a small benchmark utility for drives and enables rapid measurement of sequential and random read/write speeds. Note that CDM only supports Native Command Queuing (NCQ) with a queue depth of 32 (as noted) and shows the highest score of five runs.

CystalDiskmark – HP SDD EX920 M.2 1TB:

Benchmark Results: The HP SSD EX920 M.2 1TB drive topped out at 3210 MB/s read and 1752 MB/s write in the standard sequential write test that is done at QD32. Random 4K QD1 performance was 69.22 MB/s read and 173.0 MB/s write. Those 4K random performance numbers improved up to 414 MB/s read and 287 MB/s write at a queue depth of 32.

Manually running the sequential performance test at Q1T1 showed performance of 2025 MB/s read and 1458 MB/s write, which is mighty impressive. The sequential read performance is faster at QD1 than QD32. We also ran Random 4K numbers are a queue depth of 2 and 4 with one worker and then again at 4 with two workers for those that would like to see more data from the most important queue depth levels.

Let’s look at some other benchmarks!