PC storage has advanced in exciting new directions in recent years. PC users now have a wide range of options from standard SATA SSDs and hybrid drives to the latest PCIe 4.0 NVMe storage devices and new technologies like Intel's Optane.
Unfortunately, many of the tools for measuring storage performance were developed when HDDs were the most common type of drive. And results from synthetic benchmarks can be hard to relate to real-world performance.
PCMark 10 introduces a set of four storage benchmarks that use relevant real-world traces from popular applications and common tasks to fully test the performance of the latest modern drives.
|PCMark 10 Storage benchmark||Purpose|
|Full System Drive Benchmark||A wide-ranging test for modern drives|
|Quick System Drive Benchmark||A shorter test with lighter workloads for smaller drives|
|Data Drive Benchmark||A test for data drives and external storage devices.|
|Drive Performance Consistency Test||An extremely demanding test for expert users|
These benchmarks are for professional users in the media and industry looking for superior tools to test and compare the performance of the latest internal and external storage devices for desktop and laptop PCs.
Definition of terms used in this guide:
- An I/O or operation is the smallest piece of benchmark execution. It can be a read, write, create, close or flush.
- A trace consists of many operations. Trace defines all the operations, their order and timing.
- A pass is a series of traces, executed one at a time.
- A phase is a series of passes, run one at a time.
- A complete benchmark run is a series of phases, one or more.
Effect on device lifetime
The number of bytes written to the drive during testing depends on the test. With default settings, we have measured the following:
|Benchmark||Approx. bytes written to the drive|
|Full System Drive benchmark||204 GB|
|Quick System Drive benchmark||23 GB|
|Data Drive benchmark||15 GB|
|Drive Performance Consistency Test||23 TB plus up to 3 times the drive capacity|
The expected useful life of some storage devices is based on the number of write-erase cycles they perform. Running storage benchmarks repeatedly may shorten the life time of some drives. This is especially true of the long-running and extremely demanding PCMark 10 Drive Performance Consistency Test.