System Configuration & Testing
CPU: Intel Core i7-8700K
Motherboard: ASUS ROG Maximus X Apex
RAM: 16GB G.Skill TridentZ RGB 3200Mhz
GPU: Nvidia RTX 2080 FE
Drive: Gigabyte M.2 SSD 256GB
OS: Windows 10 Professional X64
A HDD/SSD utility software which supports a part of USB, Intel RAID and NVMe.
CrystalDiskInfo shows the Drive running on a PCIx 3.0 x2 bus with NVMe Express 1.2.
ATTO Disk Benchmark
“As the industry's leading provider of high-performance storage & network connectivity products, ATTO has created a widely-accepted Disk Benchmark freeware utility to help measure storage system performance. As one of the top tools utilized in the industry, Disk Benchmark identifies performance in hard drives, solid state drives, RAID arrays as well as connections to storage. Top drive manufacturers, like Hitachi, build and test every drive using the ATTO Disk Benchmark"
Atto gives us a bullseye on the rated 1200MB/s reads and 800MB/s writes, off to a good start!
“CrystalDiskMark is designed to quickly test the performance of your hard drives. Currently, the program allows measuring sequential and random read/write speeds."
CrystalDiskMark gives us the rated sequential speeds as expected, but manages some pretty nice 4K scores as well, even more so on the write side. This is likely thanks to the Host Memory Buffer more so that the drive.
Anvil's Storage Utilities
Anvil's Storage Utilities is a powerful tool that was designed in order to provide you with a simple means of assessing the read and write performance of your Solid State Drive or Hard Disk Drive.
The benchmark tool helps you monitor and check the response time of your unit as well as view the system information collected using Windows Management Instrumentation (WMI).
Anvil’s Storage Utilities has been designed to be the most comprehensive benchmark tool for Solid State Drives on the market. That does not mean that you can’t use it for conventional platter-based drives as well, but you do get the most out of the program when you test the speed and performance of SSDs with it. The program has gone through a series of beta and release candidate builds already and is currently available as Release Candidate 3. It is usually linked from this forum thread, which currently is not the case for the RC build which you can download here directly. You can’t use previous builds as they are set to expire automatically.
You can run a benchmark right away by selecting a drive from the menu at the top right or check out the settings first to make sure everything is configured correctly here. Here you can simulate a compression level of the test file, e.g. database, application or uncompressed, and whether you’d like to use the same test file on consecutive runs or generate a new one every time. As far as benchmarks go, you can run a standard SSD benchmark that is testing read, write, or both performances of the drive, or run threaded IO benchmarks only.
Anvil shows a little low on the sequential speeds but pretty close to the 80K read and 150K write IOPS specified.
The AS SSD benchmark determines the performance of Solid State Drives (SSD). The tool contains six synthetic and three copy tests.
The synthetic tests determine the sequential and random read and write performance of the SSD. These tests are carried out without using the operating system cache. In Seq-test the program measures how long it takes to read a 1 GB file to write respectively. 4K test the read and write performance is determined at random 4K blocks. The 4K-64 corresponds to the test Third 4K procedure except that the read and write operations are distributed to 64 threads. This test should SSDs pose with Native Command Queuing (NCQ), differences between the IDE operation mode where NCQ is not supported, and the AHCI mode. The additional compression test can measure the power of the SSD in response to the compressibility of the data. This is especially for the controllers that use to increase the performance and life of the cell compression, important.
In the first three synthetic tests and the compression test, the size of the test file 1 GB. Finally, the access time of the SSD is calculated, wherein the access to read over the entire capacity of the SSD (Full Stroke) is determined. The write access test, however, is done with a 1 GB big test file.
AS SSD uses a little different testing method but shows really close to rated speeds.
The copy benchmarks actually come in quite a bit faster than we expected.
On the compression segment, we see sustained read performance reasonably close to rated read speeds. Write performance is close as well, but we see what we would take as cache cycling if we didn’t know better. It appears our Host Memory Buffer mapping table could stand to be a just a bit bigger although most users probably won’t be slamming the drive with sustained writes very often.