To test the read / write speed of a disk, run a command prompt as administrator. To do this, in Windows, click the "Start" button, in the list of programs, find the "System Tools - Windows" section and the "Command Prompt" item in it. Right-click on it, in the context menu that opens, select the "Advanced\u003e Run as administrator" sub-item and confirm the action.
To run the predefined test suite, type the command:
Winsat disk
and press the "Enter" key. When this command completes, the speed test results will be displayed in the console window.
We are interested in the following lines:
- Disk Random 16.0 Read - read speed of random 256 blocks of 16 KB (i.e. 4 MB of test data);
- Disk Sequential 64.0 Read - read speed of 256 sequential 64 KB blocks (16 MB of test data);
- Disk Sequential 64.0 Write - write speed of 256 sequential 64 KB blocks (16 MB of test data).
Also, next to the speed results, you can see the performance index of your disks, the same one that is displayed in the Windows 7 system properties window. For Windows 7 it will be in the range from 1.0 to 7.9, and for Windows 8 and 10 - from 1.0 to 9.9.
Optionally, for more sophisticated testing, you can add the following parameters to the "winsat disk" command:
- "-seq" or "-ran" - sequential or random read / write;
- "-read" or "-write" - read or write;
- "-n N" is the physical disk number (N is a number). By default, the number is "0". Cannot be used in conjunction with the "-drive" option;
- "-drive X" - drive (X is a letter without a colon). By default, the "C:" drive is tested. Cannot be used in conjunction with the "-n" option;
- "-count N" is the number of write / read test iterations, where N is a number from 1 to 50 (default is 1);
- "-iocount N" - the number of test blocks that will be written / read during testing, where N is a number from 256 to 5000 (256 by default);
- "-seqsize N" - size of blocks in bytes for testing sequential write / read speed, where N is a number from 65536 to 1048576 (default 65536);
- "-ransize N" is the block size in bytes for testing the random write / read speed, where N is a number from 16384 to 1048576 (16384 by default).
Other parameters for the "winsat disk" command are available on Microsoft TechNet (in English).
Examples of using
The following command will run the predefined set of tests for read / write speed on the "D:" drive:
Winsat disk -drive d
Test speed of reading sequential blocks on disk "D:":
Winsat disk -seq -read -drive d
Test the speed of writing random blocks on disk "D:":
Winsat disk -ran -write -drive d
Two iterations of the test for reading speed of 512 random 1MB blocks (1048576 b) on the "D:" disk (i.e., in total, 2 × 512 × 1048576 \u003d 1073741824 b \u003d 1 GB will be read):
Winsat disk -ran -read -drive d -count 2 -iocount 512 -ransize 1048576
Removable media test
Using the "winsat disk" command, you can also check the speed of removable media such as flash drives, memory cards, etc. We do not recommend performing the pre-installed set of tests with the command "winsat disk -drive X", since flash drives and memory cards work slower than the computer's hard disk and the predefined set of tests can take a very long time. It will be more efficient to use the following commands:
Winsat disk -seq -read -drive X winsat disk -seq -write -drive X winsat disk -ran -read -drive X winsat disk -ran -write -drive X
where X is the letter of the removable drive.
You can also refine the data size and other test parameters using the parameters listed above.
Whatever speed the manufacturer specifies in the characteristics of their SSDs, the user always wants to check everything in practice. But it is impossible to find out how close the drive's speed is to the declared one without the help of third-party programs. The most that can be done is to compare how quickly files are copied on a solid state drive with similar results from a magnetic drive. In order to find out the real speed, you need to use a special utility.
Solid State Drive Speed \u200b\u200bTest
As a solution, let's choose a simple program called. It has a Russified interface and is very easy to use. So let's get started.
Immediately after launch, the main window will open in front of us, which contains all the necessary settings and information.
Before starting the test, let's set a couple of parameters: the number of checks and the file size. The measurement accuracy will depend on the first parameter. By and large, the five default checks are sufficient to get correct measurements. But if you want to get more accurate information, then you can set the maximum value.
The second parameter is the size of the file, which will be read and written during the tests. The value of this parameter will also affect both the measurement accuracy and the test execution time. However, in order not to shorten the service life of the SSD, you can set the value of this parameter to 100 Megabytes.
After setting all the parameters, proceed to the selection of the disk. Everything is simple here, open the list and select our solid state drive.
Now you can go directly to testing. The CrystalDiskMark application provides five tests:
- Seq Q32T1 - testing of sequential writing / reading of a file with a depth of 32 per stream;
- 4K Q32T1 - testing of random writing / reading of 4 Kilobyte blocks with a depth of 32 per stream;
- Seq - testing sequential write / read with a depth of 1;
- 4K - testing random write / read with depth 1.
Each of the tests can be run separately, just click on the green button of the required test and wait for the result.
You can also do full testing by clicking on the All button.
In order to get more accurate results, it is necessary to close all (if possible) active programs (especially torrents), and it is also desirable that the disk is no more than half full.
Since in everyday use of a personal computer, a random method of reading / writing data is most often used (80%), we will be more interested in the results of the second (4K Q32t1) and fourth (4K) tests.
Now let's analyze the results of our test. A 128 GB ADATA SP900 disk was used as a "test subject". As a result, we got the following:
- with the sequential method, the drive reads data at a speed 210-219 Mbps;
- recording with the same method is slowest - of all 118 Mbps;
- reading with a random method with a depth of 1 occurs at speed 20 Mbps;
- writing with a similar method - 50 Mbps;
- reading and writing with a depth of 32 - 118 Mbps and 99 Mbps, respectively.
It is worth paying attention to the fact that reading / writing is performed at high speeds only with files the size of which is equal to the size of the buffer. Those with a larger buffer will both read and copy slower.
So, with the help of a small program, we can easily estimate the speed of an SSD and compare it with the one indicated by the manufacturers. By the way, this speed is usually overestimated, but with the help of CrystalDiskMark you can find out exactly how much.
Hello! I guess it's no secret that one of the most important components inside your computer or laptop is the drive that contains the operating system. A logical consequence is the question of how to perform a hard disk speed test (or SSD, if the computer is fresher).
If your operating system is installed on a slow hard disk, it does not matter how powerful your CPU or RAM is - Windows itself and installed programs will be very reluctant to start and you will not be able to enjoy full multitasking.
In the age of the Internet, there are a lot of publications that will tell you about almost any model of the drive on sale. In addition, there are a huge number of programs for testing the speed of a hard drive, the result of which will be an understanding of what your drive is capable of.
There are many paid utilities such as PCMark or PassMark that can test the entire system and quite often they can be found in tests from well-known publications. We're taking a different path and I'm going to show you four free ways to test the speed of your hard drive or solid state drive.
The real performance of an HDD or SSD in a Windows environment (and not only) is determined not only by the rotation speed of the magnetic disk or the memory of the drive chips, but also by many other important factors. The drive controller, the SATA version on the motherboard, the driver of the controller itself, the mode of operation (ACHI or IDE) - all this affects the performance of the disk subsystem (even CPU or RAM can affect performance)
Method 1. CrystalDiskMark is our main tool
Probably the most popular hard drive speed test tool is CrystalDiskMark. Almost no drive testing is complete without this utility - this situation will help you compare your results and draw the right conclusions. A big plus is the ability of the program to test not only HDD / SSD, but also flash drives and other storage media.
The application has both a distribution kit and a portable version that does not require installation. Download as usual on the official website (as always, I recommend portable).
Working with CrystalDiskMark is outrageously easy. We launch the utility, select the size of the test block (in the picture below we have chosen 1 GB), the number of test repetitions (I chose 5 - the more repetitions, the more accurate the result) and the drive itself. Press the "all" button and wait for the program to run all tests (by the way, you can run a separate test for each mode).
The screenshot on the left is the SSD speed test, and on the right is the HDD. Just so you know how big the difference is between them and how much performance gain you will get by replacing only one component in the system
Method 2. CrystalDiskInfo - detailed information about the HDD / SSD drive
At the very beginning of the post, I already wrote that the test of the speed of a hard disk or SSD will not be entirely correct if we do not find out the factors affecting the performance of the disk subsystem. The CrystalDiskInfo utility will tell you a lot of interesting things about your drive, but we are only interested in one nuance - download the application from the official website and run it.
Pay attention to the line "Transfer mode", in the picture below I have this (SATA / 600 | SATA / 600). These parameters must match, i.e. by connecting the SSD drive to the SATA / 300 port (this is the SATA II standard), we will get the maximum exchange rate with the disk of 300 MB, and if we look at the performance test in the first method, we see that the maximum read speed was far beyond 300 ...
By connecting such a high-speed drive to a SATA or SATA II port, its performance will simply rest on the performance of the controller (with classic HDDs it is not so critical, since even SATA capabilities are abundant)
In general, CrystalDiskInfo can tell you about temperature, drive time and many other useful indicators. For owners of classic HDD, the Reallocate Sector item will be useful - thanks to it, you can predict the failure of the device
Method 3. AS SSD Benchmark - a healthy competitor to CrystalDisk from the Germans
The Germans know how to make not only adult films, but also excellent utilities for testing the speed of a hard drive or SSD. In this case, I want to introduce you to the AS SSD Benchmark application, the functionality of which is very similar to CrystalDiskMark, but in contrast to it, it also shows the access time to data (and in general there are still minor differences).
You can download from the official website (it's in German, the download link is at the end of the page), the application itself is in English (many bloggers have a version exclusively in German)
The utility is portable and does not require installation, just run the application, mark the necessary tests and press START, everything is the same as in the first method. On the left is my home SSD, on the right is a classic HDD.
Please note that the TOOLS menu has a couple of interesting tests that can predict the performance of the drive when copying ISO files, programs or various toys - CrystalDiskMark does not have such functionality.
Method 4. HD Tune is a good tool with visual graphics
HD Tune is probably the most famous hard drive speed test application, but it comes in last place in today's rankings for a reason. The fact is that the free version of HD Tune has not been updated since February 2008 ... however, it still works in 2k17 on the newest Windows 10. Download, as always, from the official website (unfortunately, there is no portable version)
After passing the test, a visual reading graph will be available to us (along with the maximum and minimum values, as well as the speed of data access). In general, the information is useful, but there is no way to test the write speed on the disk, which is a little upsetting ...
In view of its antiquities the application may incorrectly detect modern drives, but this does not affect the test results
Conclusion about programs for testing hard disk speed
It's time to draw conclusions. We tested the speed of a hard disk or SSD using four different programs (or rather, there are only three applications for testing, and one more utility to be sure that the tests will be objective).
In reality, there are many more programs that allow you to check the speed of a hard drive, but I decided to introduce you to the leaders of this niche ... but if you have anything to add, I’m waiting for you in the comments.
It is believed that one of the most significant disadvantages of solid-state drives is their ultimate and, moreover, relatively low reliability. Indeed, due to the limited resource of flash memory, which is caused by the gradual degradation of its semiconductor structure, any SSD sooner or later loses its ability to store information. The question of when this can happen remains a key question for many users, so many buyers, when choosing drives, are guided not so much by their performance as by reliability indicators. Manufacturers themselves also add fuel to the fire of doubts, who, for marketing reasons, under the terms of a guarantee for their consumer products, stipulate relatively low volumes of allowed recording.
In practice, however, mainstream solid-state drives are more than reliable enough to be trusted to store user data. An experiment that showed the absence of real reasons for worrying about the finiteness of their resource was carried out some time ago by the TechReport website. He performed a test that showed that, despite all the doubts, the endurance of the SSD has already grown so much that you can not think about it at all. As part of the experiment, it was practically confirmed that most models of consumer drives are capable of transferring the recording of about 1 PB of information to their failure, and especially successful models, like the Samsung 840 Pro, survive, digesting even 2 PB of data. Such recording volumes are practically unattainable in the conditions of a conventional personal computer, therefore, the life of a solid-state drive simply cannot come to an end before it is completely obsolete and replaced by a new model.
However, this testing could not convince skeptics. The fact is that it was carried out in 2013-2014, when solid-state drives built on the basis of planar MLC NAND, which is manufactured using a 25-nm process technology, were in use. Before its degradation, such memory is capable of transferring about 3000-5000 programming-erasure cycles, and now completely different technologies are in use. Today, flash memory with a three-bit cell has come to mass models of SSD, and modern planar technical processes use a resolution of 15-16 nm. In parallel, flash memory with a fundamentally new three-dimensional structure is gaining popularity. Any of these factors can radically change the situation with reliability, and in total, modern flash memory promises only a resource of 500-1500 rewriting cycles. Is it possible that along with memory, drives are also deteriorating and you need to start worrying about their reliability again?
Most probably not. The fact is that along with the change in semiconductor technologies, there is a continuous improvement of the controllers that control flash memory. They are introducing more advanced algorithms that should compensate for the changes taking place in NAND. And, as manufacturers promise, current SSD models are at least as reliable as their predecessors. But the objective ground for doubts still remains. Indeed, on a psychological level, drives based on the old 25nm MLC NAND with 3000 rewrite cycles look much more solid than modern SSD models with 15 / 16nm TLC NAND, which, other things being equal, can only guarantee 500 rewrite cycles. The growing popularity of TLC 3D NAND, which, although produced according to larger technological standards, is not too encouraging, is also subject to a stronger mutual influence of cells.
Taking all this into account, we decided to conduct our own experiment, which would allow us to determine what endurance can be guaranteed by current drive models based on the currently most popular types of flash memory.
Controllers decide
The finiteness of the life of drives built on flash memory has been no surprise to anyone for a long time. Everyone has long been accustomed to the fact that one of the characteristics of NAND memory is a guaranteed number of rewriting cycles, after exceeding which cells can begin to distort information or simply fail. This is explained by the very principle of operation of such a memory, which is based on the capture of electrons and the storage of the charge inside the floating gate. The change in the state of the cells occurs due to the application of relatively high voltages to the floating gate, due to which the electrons overcome the thin dielectric layer in one direction or the other and are retained in the cell.
Semiconductor NAND Cell Structure
However, such a movement of electrons is akin to a breakdown - it gradually wears out the insulating material, and ultimately this leads to a violation of the entire semiconductor structure. In addition, there is a second problem, which entails a gradual deterioration of the characteristics of the cells - when tunneling occurs, electrons can get stuck in the dielectric layer, preventing the correct recognition of the charge stored in the floating gate. All this means that the moment when the cells of flash memory stop working normally is inevitable. New technological processes only exacerbate the problem: the dielectric layer only becomes thinner with decreasing production standards, which reduces its resistance to negative influences.
However, it would not be entirely correct to say that there is a direct relationship between the resource of flash memory cells and the lifespan of modern SSDs. The work of a solid state drive is not a straightforward write and read in flash memory cells. The fact is that NAND memory has a rather complex organization and special approaches are required to interact with it. Cells are grouped into pages and pages are grouped into blocks. Writing data is possible only in blank pages, but in order to clear the page, you must reset the entire block. This means that writing, and even worse, changing data, turns into a complex multi-step process, including reading a page, changing it and rewriting it to free space, which must be cleared beforehand. Moreover, the preparation of free space is a separate headache that requires "garbage collection" - the formation and cleaning of blocks from pages that have already been used, but have become irrelevant.
The diagram of the operation of flash memory of a solid state drive
As a result, the actual volumes of writing to the flash memory can differ significantly from the volume of operations initiated by the user. For example, changing even one byte may entail not only writing an entire page, but even the need to rewrite several pages at once in order to pre-free a clean block.
The relationship between the amount of writing the user makes and the actual load on the flash memory is called the write gain. This coefficient is almost always higher than one, and in some cases - much. However, modern controllers have learned to effectively reduce the write gain due to buffering operations and other intelligent approaches. Technologies such as SLC caching and wear leveling, useful for extending the life of cells, have become widespread. On the one hand, they put a small portion of the memory in a sparing SLC mode and use it to consolidate small disparate operations. On the other hand, they make the load on the memory array more even, preventing unnecessary multiple overwrites of the same area. As a result, storing the same amount of user data on two different drives from the point of view of the flash memory array can cause completely different load - it all depends on the algorithms used by the controller and the firmware in each case.
There is another side: garbage collection and TRIM technologies, which, in order to improve performance, pre-prepare clean blocks of flash memory pages and therefore can transfer data from place to place without any user intervention, make an additional and significant contribution to the wear of the NAND array. ... But the specific implementation of these technologies also largely depends on the controller, so the differences in how SSDs manage their own flash memory resource can be significant here as well.
As a result, all this means that the practical reliability of two different drives with the same flash memory can differ very noticeably only due to different internal algorithms and optimizations. Therefore, speaking about the resource of a modern SSD, you need to understand that this parameter is determined not only and not so much by the endurance of memory cells, but by how carefully the controller handles them.
The algorithms for SSD controllers are constantly being improved. The developers are not only trying to optimize the amount of write operations in flash memory, but are also introducing more efficient methods of digital signal processing and correction of reading errors. In addition, some of them resort to the allocation of a large spare area on the SSD, due to which the load on the NAND cells is further reduced. All this also affects the resource. Thus, SSD manufacturers have a lot of leverage in their hands to influence how much endurance their product will demonstrate, and the resource of flash memory is just one of the parameters in this equation. That is why endurance tests of modern SSDs are of such interest: despite the widespread introduction of NAND memory with relatively low endurance, current models do not necessarily have to be less reliable than their predecessors. Advances in controllers and the way they operate are quite capable of compensating for the flimsy modern flash memory. And this is precisely why research on current consumer SSDs is interesting. Compared to previous generations of SSDs, only one thing remains unchanged: the resource of solid-state drives is in any case finite. But how it has changed in recent years is exactly what our testing should show.
Testing technique
The essence of SSD endurance testing is very simple: you need to continuously overwrite data in drives, trying to set the limit of their endurance in practice. However, a simple linear recording does not quite meet the test objectives. In the previous section, we talked about the fact that modern drives have a whole bunch of technologies aimed at reducing the write gain, and in addition, they perform garbage collection and wear leveling procedures differently, and also react differently to the command of the TRIM operating system. ... That is why the most correct approach is to interact with the SSD through the file system with an approximate repetition of the profile of real operations. Only in this case will we be able to get a result that ordinary users can consider as a guideline.
Therefore, in our endurance test, we use drives formatted with the NTFS file system, on which files of two types are continuously and alternately created: small - with a random size of 1 to 128 KB and large - with a random size from 128 KB to 10 MB. During the test, these files with random filling multiply until there is more than 12 GB of free space on the drive, but when this threshold is reached, all created files are deleted, a short pause is made, and the process is repeated again. In addition, the tested drives also contain the third type of files - permanent. Such files with a total volume of 16 GB do not participate in the erasure-rewriting process, but are used to check the correct operability of the drives and the stable readability of the stored information: each cycle of filling the SSD, we check the checksum of these files and compare it with the reference, pre-calculated value.
The described test scenario is reproduced by a special program Anvil's Storage Utilities version 1.1.0, the status of drives is monitored using the CrystalDiskInfo utility version 7.0.2. The test system is a computer with an ASUS B150M Pro Gaming motherboard, a Core i5-6600 processor with integrated Intel HD Graphics 530, and 8GB DDR4-2133 SDRAM. SATA drives connect to the SATA 6Gb / s controller built into the motherboard chipset and operate in AHCI mode. The Intel Rapid Storage Technology (RST) driver 14.8.0.1042 is used.
The list of SSD models taking part in our experiment currently includes more than five dozen items:
- (AGAMMIXS11-240GT-C, firmware SVN139B);
- ADATA XPG SX950 (ASX950SS-240GM-C, firmware Q0125A);
- ADATA Ultimate SU700 256 GB (ASU700SS-256GT-C, firmware B170428a);
- (ASU800SS-256GT-C, firmware P0801A);
- (ASU900SS-512GM-C, firmware P1026A);
- Crucial BX500 240GB (CT240BX500SSD1, firmware M6CR013)
- Crucial MX300 275GB (CT275MX300SSD1, firmware M0CR021)
- (CT250MX500SSD1, firmware M3CR010);
- GOODRAM CX300 240 GB ( SSDPR-CX300-240, firmware SBFM71.0);
- (SSDPR-IRIDPRO-240, firmware SAFM22.3);
- (SSDPED1D280GAX1, firmware E2010325);
- (SSDSC2KW256G8, firmware LHF002C);
Good day.
The speed of the entire computer as a whole depends on the speed of the disk! Moreover, surprisingly, many users underestimate this moment ... But the speed of loading Windows OS, the speed of copying files to / from disk, the speed of launching (loading) programs, etc. - everything depends on the speed of the disk.
Now in PCs (laptops) there are two types of disks: HDD (hard disk drive - familiar hard drives) and SSD (solid-state drive - newfangled solid-state drive). Sometimes their speed differs significantly (for example, Windows 8 on my computer with an SSD starts in 7-8 seconds, versus 40 seconds with an HDD - the difference is enormous!).
And now about what utilities and how you can check the speed of the disk.
One of the best utilities for checking and testing disk speed (the utility supports both HDD and SSD drives). Works in all popular Windows OS: XP, 7, 8, 10 (32/64 bits). Supports Russian (although the utility is quite simple and easy to understand without knowing English).
Fig. 1. The main window of the CrystalDiskMark program
To test your disk in CrystalDiskMark you need:
- select the number of write and read cycles (in Fig. 2 this number is 5, the best option);
- 1 GiB - file size for testing (best option);
- "C: \\" is the drive letter for testing;
- to start the test just press the “All” button. By the way, in most cases they are always guided by the line "SeqQ32T1" sequential write / read - therefore, you can simply select a test of this particular option (you need to press the button of the same name).
The first speed (the Read column, from the English "read") is the speed of reading information from the disk, the second column is writing to the disk. By the way, in fig. 2 SSD disk (Silicon Power Slim S70) was tested: read speed 242.5 Mb / s - not a good indicator. For modern SSDs, the optimal speed is considered to be at least ~ 400 Mb / s, provided that it is connected via SATA3 * (although 250 Mb / s is more than the speed of a conventional HDD and the increase in speed is visible to the naked eye).
* How to determine the operating mode of a SATA hard drive?
From the link above, in addition to CrystalDiskMark, you can also download another utility - CrystalDiskInfo. This utility will show you the SMART disk, its temperature, and other parameters (in general, an excellent utility for obtaining information about the device).
After starting it, pay attention to the line "Transfer mode" (see Fig. 3). If this line displays SATA / 600 (up to 600 MB / s) - then the disk is operating in SATA 3 mode (if the line displays SATA / 300 - i.e. the maximum bandwidth of 300 MB / s is SATA 2) ...
AS SSD Benchmark
Author`s site: http://www.alex-is.de/ (download link at the very bottom of the page)
Another very interesting utility. Allows you to easily and quickly test the hard drive of your computer (laptop): quickly find out the speed of reading and writing. It does not need installation, use it as standard (as with the previous utility).
