Hard drives have run pretty well over the years. When I think about the number of laptops and PCs I’ve owned and used over the past 25 years, I can only think of one failure. In that case, all of the data was easily recoverable. There was another incident – a very early small form factor external drive from Western Digital. I loved it and stored all my files on it. It was great for backup and when I moved from one device to another. Unfortunately, it was knocked off a desk and it wouldn’t start. The good folks at Western Digital kindly agreed to let their forensic specialists try to recover my data but it was not to be.
The point is that hard drives have largely stood the reliability test of time. Where they tend to differentiate themselves, therefore, is in terms of performance and especially speed. Accordingly, there are a number of hard drive benchmarking tools out there to help you test your drives. These can be useful when debugging device performance, in drive selection, and in many other ways. It is important to pick components with matching performance categories. The last thing you want is a drive that can’t keep up with the bandwidth, latency, and communication pace of the rest of the hardware.
Hard Drive Benchmarking Tools Selection Tips
Although it might seem simple to test the performance of a hard drive there are a number of factors to consider.
- MB v Mb: There is often some confusion between megabits (Mb) and Megabytes (MB). An Mb is 1,000,000 bits. An MB is 1,000,000 bytes. Watch out then for Mbps (Megabits per second) and MBps (Megabytes per second). The former specifies internet connection speeds. MBps specifies how much of a file is downloaded/uploaded per second. In some scenarios one or other may have more weight when it comes to product selection.
- Raw speed: Some hard drives are designed to be faster than others. Some are as slow as 1200 Mbps and 150 MBps. From there they grade up to between 3000 Mbps and 375 MBps to 6000 Mbps and 750 MBps for SATA. SAS drives are faster, capable of 600MBps and even 1500MBps transfer speeds in some cases. But they are more expensive and typically contain less storage capacity.
- Make sure that you don’t confuse your metrics and that you are comparing apples to apples. If you miss the subtle difference between mbps and MBps, you could end up thinking a drive was faster when in fact it is much slower, or vice versa.
- RPM: RPM or revolutions per minute is another common metric. This is simply how fast the disk spins. Consumer-level SATA typically operates at 5400 rpm and can go as high as 7200 rpm. SAS generally operates between 7200 rpm and 15,000 rpm.
- Test design: Some tests are designed for a very specific purpose or by a vendor. They may be weighted around how a drive performs with a certain application such as SAP, or on a type of workload. Alternatively, they can be designed by a vendor to emphasize the value of its own products. Such tests are set up to show a drive in its best light, hitting the sweet spot for that particular drive. Make sure, then, that the test you are using matches the workloads, applications, and use cases you use in your environment.
- Test on the ground. Lab tests are good. They have value. But run the test in your own production environment, too. That gives you a real-world result on whether a device fits your needs or not.
Top Hard Drive Benchmark Tools and Vendors
Many different hard drive benchmarking tools are available to measure transfer rate and transfer speed under different scenarios. Here are some of the top ones.
Atto offers disk benchmarks for Windows and Mac. Its Disk Benchmark for Windows, for example, measures storage system performance. It identifies performance in hard drives, solid state drives, RAID arrays as well as the host connection to attached storage. Drive manufacturers are included in its customer base.
- ATTO Disk Benchmark for Windows can be used to test any OEM RAID controller, storage controller, host adapter, hard drive or SSD drive, and notice that ATTO products consistently provide the highest level of performance to your storage.
- Custom transfer size.
- Support for overlapped I/O.
- Supports various queue depths.
- I/O comparisons with various test patterns.
- Timed mode allows continuous testing.
- Custom transfer sizes.
- A separate Disk Benchmark for macOS is available.
Crystal Disk Mark
Crystal Disk Mark tests the reading and writing speeds of hard drives. It includes a variety of tests that allow users to check performance under different scenarios. As some drives do better on reading or writing data, it can test both aspects. Users decide the number of tests they wish to do, the size (in megabytes) of the test, and the drive.
- Reading and writing capabilities are tested sequentially in blocks of 512 kilobytes.
- Can test in blocks as small as 4 kilobytes.
- If multiple tests are done, Crystal Disk Mark calculates the average reading/writing velocity between all the tests.
- Detects possible bottlenecks.
- Measures sequential and random performance (read/write/mix).
- Can measure various modes (peak/real world/demo).
- Support for multiple languages.
MiniTool Partition Wizard Server is far more than a disk benchmarking tool. But it does that function well. In addition to testing, it makes it easier for users to redistribute disk space and ensure system integrity. Its disk/partition copy function allows users to clone the whole drive rapidly. To extend a RAID array, the extend partition feature helps enlarge hard drive space flexibly. For data safety, its power-off protection mode makes it possible to revert the operations being performed when power outage occurs.
- Benchmarking tools help keep servers running well.
- MiniTool Partition Wizard Server has can extend C drive on Windows Server and ensure data safety.
- Can change cluster size and create NTFS partitions larger than 16 TB.
- Can copy an entire disk for hard drive replacement.
- Improved FAT32 partition data recovery & BitLocker partition clone.
- Support for Windows Server 2019/2016/2012/2008 (R2)/2003 (R2).
- Also works for Windows 10/8.1/8/7/Vista/XP.
- MiniTool Partition Wizard Server is for server users to reconfigure hard drive on one computer.
- For Enterprise users, MiniTool Partition Wizard Enterprise that works for 99 computers in one company.
- MiniTool Partition Wizard Technician allows usage on 299 computers anywhere.
Although the AS SSD test is squarely aimed at solid state devices (SSDs), it is included as many users now have SSDs somewhere in their environment. Testing HDDs alone may not be enough. This free utility conducts multiple tests to reveal any problems with any SSDs operating. It can be stored on an USB for on the move testing.
- Compatible with Windows 7 and above.
- Find out the speed of all installed SSDs.
- Reveals underlying performance or other issues.
- Three separate tests to provide evidence on a driver’s general behavior while it copies, reads, and writes data.
- Determines the access time of an SSD, along with its speed and performance capabilities.
- Measures the speed at which an SSD reads and writes 1 GB of data.
- A 4K test determines the read and write speed at 4K blocks.
- Results are displayed in IOPS or MB per second.
HD Tune Pro by EDF Software is a hard disk and SSD utility with many different functions. As well as being used to measure a drive’s performance, it can also scan for errors, and check health status using Self-Monitoring, Analysis and Reporting Technology (SMART). This is a monitoring system included in hard drives and SSDs to detect and report indicators of drive reliability.
- Provides a write benchmark.
- Can securely erase all data.
- Includes a folder usage view feature.
- Includes a disk monitor and file benchmark.
- Offers command line parameters.
- Overtemperature is shown as a warning.
UserBenchmark boasts that it can speed test a PC in less than a minute. It is a small open source collective of scientists and engineers that computes and analyzes millions of benchmarks. Its website includes huge numbers of tests of SSDs, HDDs, CPUs, RAM, and more.
- No affiliation to major CPU or disk manufacturers.
- Ability to compare the results of a great many tests of HDDs to get a wider view of overall results.
- Has a freeware PC speed test tool that gives individual results and compares them to other users with the same components.
- Drive tests include: read, write, sustained write, and mixed IO.
- RAM tests include: single/multi core bandwidth and latency.
- Reports are generated and presented on userbenchmark.com.
Originally developed by Intel, Iometer was donated to the Open Source Development Lab (OSDL). It is an I/O subsystem measurement and characterization tool for single and clustered systems. It is maintained by an international group of individuals who are continuously improving, porting, and extending the product. It is particularly good at benchmarking IOPS.
- Supports Windows, Linux, and Mac.
- It is used as a benchmark and troubleshooting tool that can be configured to replicate application behavior.
- Measures the performance of disk and network controllers.
- Can measure the bandwidth and latency capabilities of buses.
- Monitors network throughput to attached drives.
- System-level hard drive performance and network performance metrics are included.
- Can generate synthetic I/O loads.
Roadkill performs disk speed tests and provides details about how fast they transfer data. This includes data transfer rate for linear reads, random read transfer rate, and the seek time of the drive. An overall score is given as a comparison with various brands of drives. Its drive testing and other applications are provided free of charge.
- Information about the drive such as firmware revision, model, size, file system and more is displayed.
- Roadkill also hosts applications provided free of charge for personal or business use.
- Runs read tests only.
- Displays a score ranging from 512 bytes to 1MB blocks.