Introduction of mainstream SSD testing Softwares

• AS SSD Benchmark

AS SSD Benchmark is a special SSD testing software from Germany, which can test the performance of continuous read and write, 4K alignment, 4KB random read and write and response time, and gives a comprehensive score.
The test includes four aspects: sequential read and write (Seq), 4K random read and write (4K), 64 thread 4K read and write (4K-64Thrd), seek time (Acc.time). A score of (Score) will be given after the test is completed.
There are two other modes to choose from, namely MB/s and IOPS!

AS SSD Benchmark will generate and write test data files of 5GB during testing, and all three test transfer rate projects read and write these data files to convert speed.
Its 4KB QD64 is mainly used to measure the gap between NCQ (Native Command Queuing, native command queues.
IDE mode is no different from ordinary 4KB random.
Because each test requires a certain size of data to read and write, the lower the performance of the hard disk, the longer the test will take. It is not appropriate to run this test with a mechanical hard disk. It will take about an hour to finish the whole run.
The addressing time test is there, where reading is the test addressing random 4KB file (full LBA area), and writing is testing addressing random 512B file (specified 1GB address range).
Note that running the AS SSD benchmark requires at least 2GB free space.

AS SSD Benchmark can not only test the performance of SSD, but also detect the firmware algorithm of SSD, whether to turn on AHCI mode, whether to align with 4K and so on. It is the most widely used SSD testing software at present.

AS SSD Benchmark test data are random.

---

• ATTO Disk Benchmark

ATTO Disk Benchmark is an easy-to-use disk transfer rate detection software, which can be used to detect the read and write rates of hard drives, USB drives, memory cards, and other removable disks.
The software uses data test packages of different sizes, and the data packets are read and written at 0.5K, 1.0K, and 2.0K until they are read and written to 8192.0KB respectively. After the test is completed, the data is expressed in the form of a bar chart, which shows the effect of different file size ratios on disk speed.

The ATTO test is the disk continuous read and write performance in the limit case, and the test model adopted has high compressibility. ATTO tests all zero data by default.

---

• CrystalDiskMark

CrystalDiskMark software is a small tool for testing your hard drive or storage device, the size of the test storage device, and the number of tests that can be selected.
The test project is divided into continuous transfer rate test (block unit 1024KB), random 512KB transmission rate test, random 4KB test, random 4KB QD32 (queue depth 32) test.
CrystalDiskMark runs 5 times by default, and the data volume of each 1000MB is the best.
Before testing the CrystalDiskMark software, a test file is also generated (the size depends on the user settings).
In general, the larger the setting, the less interference the data cache will cause, and the performance will better reflect the real performance of SSD, but the disadvantage is that it will seriously affect the durability of SSD (writing too much data will affect SSD E).
Therefore, software default values are generally adopted in testing.

The default data of the software is incompressible data.
If < All 0x00, 0Fill > or < All 1 × 00.1Fill > is selected in the setting options, the test results will be very different.
In fact, it is to change the data model to all compressible continuous data, which is the same as the ATTO test principle, the test results are quite good, but the actual reference significance is not great.
After modifying the data model, there is an obvious feature that will be marked directly on the title bar of the CDM.

---

• PCMark Vantage

PCMark Vantage can measure the overall performance of various types of PC.
From multimedia home entertainment systems to notebooks, from professional work stations to high-end game platforms, whether in the hands of professionals or ordinary users, can be thoroughly understood in PCMark Vantage, so as to maximize performance.
The test can be divided into the following three parts:

1. Processor Testing: based on data encryption, decryption, compression, decompression, graphics processing, audio and video transcoding, text editing, web rendering, mail functions, processor artificial intelligence game testing, contact creation, and search.
2. Graphics Testing: based on HD video playback, graphics card graphics processing, game testing.
3. Hard Disk Testing: use Windows Defender, “Alan Wake” games, image import, Windows Vista startup, video editing, media center use, Windows Media Player search and classification, and launch: Office Word 2007, Adobe Photoshop CS2, Internet Explorer, Outlook 2007 of the following programs.

---

• IOMeter

Iometer is a measurement and description tool for the Icano subsystem of a stand-alone or cluster.
Compared with the previous test software, Iometer is relatively free in the test software. Users can configure the test disk data range, queue depth, data mode (compressible or incompressible, some versions support, some older versions do not support), test mode (random or sequential access), read-write test ratio, random and sequential access ratio, test time, and so on.

Local IO performance Testing

1. Start EXE, and click the Iometer icon on windows;
2. Select a drive in the Disk Targets page;
3. Select a desired test project on the Access Specifications page;
4. On the “Results Display” page, set the “Update Frequency (Seconds)” setting to count the test results. If you do not set the setting, not only the test results are not displayed during the test, but also there is no data in the test results file after the test is over.
   • Total I/Os Per Second: Data access speed. The higher the value is, the better it is.
   • Total MBs Per Second: Data transfer speed. The higher the value is, the better it is.
   • Total MBs Per Second: Data transfer speed. The higher the value is, the better it is.
   • Average I/O Response Time: The smaller the value is, the better it is.
   • CPU Utilization: The lower the value is, the better it is.
5. Click the “Start Tests” button in the toolbar, select a test results output file and start a test (usually a test runs for 10 minutes);
6. Click the button to stop all tests when the test is completed.
7. View test results, because IOMeter does not provide a tool for GUI to view test reports. Can use EXCEL to open the test results files “CSV”, and then use EXCEL’s icon tool to organize the test results. Or use the “Import Vizard for MS Access” provided by IOMeter to import the test results into an Access file.

---

References:

1.http://diy.pconline.com.cn/319/3190289.html

2.http://www.ssdfans.com/?p=11