Page 8 - Benchmark: PassMark PerformanceTest 8.0
About PassMark PerformanceTest 8.0
This Advanced Disk Test, which is part of PerformanceTest, measures the data transfer speed when reading or writing data to one or more disks. The speed that data can be transferred between memory and a hard disk drive is one of a system's most important performance aspects. There are quite a few factors which have a bearing on this speed and the Advanced Disk Drive Test allows the user to vary most of these factors and compare the results.
The test supports any drive that can be mounted under Windows. Including IDE drives, SCSI, RAID, USB key drives, SATA, networked shared drives and external drives.
Users have the ability to test multiple drives at the same time using multiple threads, and specify:
- The size of the test file used. Larger files mean that the system cache has less of an effect on the test types, which use caching (see below).
- The size of the data block used for each read or write request. Larger blocks mean less requests and can lead to an improvement in performance.
- The choice of four access methods - C/C++ API, Win32 API cached / uncached and raw disk access.
- Sequential or random access (seeking plus reading and writing)
- Synchronous and Asynchronous access
- The split between reading and writing
The results of all completed tests may be graphed using our custom graphing components.
From: Developer's Page
PassMark PerformanceTest 8.0's Advanced Disk Test, unlike HD Tune Pro 4.60, generates some awesome graphs right out of the box. It also provides valuable insight in simulating real world performance applications. To make things clear to you, the first graph simulates a database server, followed by a file server, web server, and workstation. Obviously, PassMark PerformanceTest 8.0 uses highly compressible data in most tests some controllers can really take advantage of. However, it also requires high IOPS capabilities for the best score. In the past, SandForce, Marvell, and Indilinx based drives all perform pretty well in this test. One thing clear is the Kingston HyperX Predator PCIe 480GB provided reasonably flat graphs, which is good if you are looking for consistent performance.
Overall, this M.2 PCIe solid state drive's performance was a league of its own in three out of four categories. With results of 18.24MB/s, 566.66MB/s, 1059MB/s, and 117.52MB/s, all are amazing results for a single drive configuration, save for the first test. At 18.24MB/s, the HyperX Predator really fell behind in the database simulation. Almost everything we have tested in the past was several times faster. To make sure it was not a one time thing, I retested the SSD several times, only to come out with the same results. Compared against the OCZ RevoDrive 350 480GB, the corresponding numbers were 182.65MB/s, 1084MB/s, 1062MB/s, and 94.53MB/s, respectively. In case you are curious, if we look at the OCZ Vector 150 240GB, one of the best single drives in the segment (Which is actually Indilinx Barefoot 3, by the way), it delivered 142.09MB/s, 426.34MB/s, 475.82MB/s, and 73.10MB/s, respectively. Where the Kingston HyperX Predator PCIe 480GB really fell behind in the database run, it completely destroyed everything else by the massive 1059MB/s score in the web server benchmark.
Page Index
1. Introduction, Packaging, Specifications
2. A Closer Look, Test System
3. Benchmark: AIDA64 Disk Benchmark
4. Benchmark: ATTO Disk Benchmark
5. Benchmark: Crystal Disk Mark 3.0
6. Benchmark: HD Tach 3.0.1.0
7. Benchmark: HD Tune Pro 4.60
8. Benchmark: PassMark PerformanceTest 8.0
9. Benchmark: PCMark Vantage
10. Conclusion
