Real-world disk benchmarking with Disk Bench

Test your hard disks under real-world conditions with this free benchmarking utility.

Benchmarking your hard disks tells you how fast they transfer data. There's no shortage of free benchmarking tools, but most of them run a single specific test that often doesn't tell you much about how your disks perform in a real-world situation, such as copying large files and even whole directories. Disk Bench is a free benchmarking tool that lets you choose which files or folders to copy. It then copies the data, benchmarks the speed, and deletes the copied data from the destination, though you can also keep the copied data. It also creates test files with a repeating 128-byte string, which is a preferred method for benchmarking a single disk, as well as batch files for testing.

Disk Bench's compact interface is divided horizontally between the tool's entry fields and controls on top and its display window below. The tabs describe each function: Copy File, Copy Directory, Create File, Read File, and Create Batch File. We started with the Copy File tool, which is quite simple but a good choice for just the sort of real-world benchmarking Disk Bench is designed to extract. We browsed to a Source File on our C drive, and then browsed to a destination on our D drive. By default, Disk Bench selects the check box labeled Kill Destination File, which automatically deletes the copied data once the test is finished. We clicked Start Bench, and our test finished with lightning speed, showing nearly 200MB/s speeds for our C drive, a SATA II SSD.

The Create File tab's tests proved more demanding. This tab has two Destination File fields, though one of them is inactive until you select the check box to create two threaded files. Blocks can be in KB, MB, and GB size, though obviously your system's memory resources will be the upper limit, as we found out with gigabyte-size blocks. Once or twice on large tests, Windows showed Disk Bench as Not Responding, yet it finished the test each time, except for one rather large file, and no shame there. These tests yielded much more realistic benchmarks in the 100MB/s range. This compact freeware belongs in any benchmarking kit.

CNET Top 5
Companies Apple could buy with their billions
Apple's sitting on a massive pile of cash. Here are five interesting ways they could spend it.
Play Video

Member Comments