Computer system unit, device and composition
So, what does the computer system unit consist of? An important clarification here is that under the word “computer” different people can mean, oddly enough, different things: a monitor, a…

Continue reading →

Computer motherboard, what is it for and what does it look like
The computer motherboard is that foundation on which all the components of the system unit are built. The role of the computer motherboard cannot be overestimated. After all, it depends…

Continue reading →

How to make a bootable USB flash drive Windows 7 and Windows 8
A multiboot flash drive that has several operating systems on board at once, for example, Windows XP, Windows 7 and Windows 8, is a very useful tool that any modern…

Continue reading →

What are SSD drives and how do they work?

Today we will analyze the main points and principles of functioning of the technology of solid-state SSD drives. As you remember, in the first part of the article we conducted a comparative test of one SSD and two HDD drives. We examined how it looks from the inside and of what basic blocks it consists of.

Also – they listed the main advantages of this technology, and now we will consider the disadvantages that are inherent in it at the moment. We will present the main ones in the form of a list:

High (relative to HDD disks) cost of data storage, i.e. – we get less disk capacity for big money
Great vulnerability (with respect to devices with magnetic recording principle) to electrical interference and problems of power supply (sudden power outage, magnetic fields, static electricity)
It is impossible to completely fill the disk (15-20% of the space should be free)
The life of the media is limited by a certain number of write cycles of its cells.
But let’s – in order! To begin with, what is an SSD drive and what is the principle of its operation?

SSD driveSSD drive

This is a solid-state drive in which NAND flash memory chips are used instead of traditional hard drive plates coated with a ferromagnetic layer.

NAND memory chips NAND memory chips

NAND memory is an evolution of flash memory, the chips of which had much lower performance, durability and structurally looked more massive.

You may be interested in the fact that flash memory was developed in one of Toshiba’s divisions in 1984. The first commercial chip based on this development in 1988 was released by Intel. And a year later (in 1989), the same Toshiba introduced a new type of flash memory – NAND.

At the moment there are three main options (modifications) of NAND memory:

SLC (Single Level Cell)
MLC (two-level – Multi Level Cell)
TLC (Three-Level – Three Level Cell)
The most expensive and reliable solutions are devices based on SLC chips. Why? They allow you to store only one bit of information in each memory cell. Unlike them, MLC and TLC chips can store two and three bits, respectively. This was made possible through the use of different levels of electric charge on the gates of the memory cells.

Schematically, this can be represented as follows:

Such a multi-level structure allows you to dramatically increase the capacity of the chips with the same physical volume (as a result, each gigabyte is cheaper). BUT! Nothing is free! Therefore, the MLC and TLC chips sharply reduced their “life”, which is directly related to the number of cycles of rewriting their cells.

For SLC, this is 100,000 erase / write cycles, for MLC – 10,000, and for TLC – only 5,000. This decrease in reliability is associated with the gradual destruction of the dielectric layer of the floating gate of the cell due to a small reserve of change in its state under the influence of electric current. Plus, due to the fact that with each new level the task of error-free recognition of the level of the electric signal becomes more complicated, which means that the total search time of the desired cell with data increases, the likelihood of reading errors increases.

To combat the phenomena described above, manufacturers have to develop specialized highly intelligent control microcontrollers for SSD drives, which, in addition to input-output procedures, must write information to the media so that its flash memory chips wear out evenly and control this wear, balancing the load, also – carry out error correction, etc.

It is the controller that is the weak point of solid-state SSD drives, as it is more sensitive to power problems and damage to the firmware (firmware) located in it can lead to a complete loss of all user data. And their correct restoration is an even more time-consuming operation than in the case of HDD disks. Due to the fact that the data are scattered across different memory chips and it is necessary to correctly restore their original structure, and this is not easy.

Therefore, manufacturers of SSD drives regularly update the firmware of their drives and post them for free download, finalizing and improving the algorithms of the device and preventing data loss in case of an emergency.

Recommendation! Periodically update the “firmware” of your SSD by downloading it from the manufacturer’s website.

Manufacturers also struggle with the wear and tear of MLC memory cells using a method that has proven itself in disks with a magnetic recording principle: reserving part of their volume (10-20%) for the dynamic replacement of worn cells. In the case of HDD, this area serves to replace the bad sectors of the hard drive.

But we, as users, can help our SSD drive idle without wasting our limited “life” resource and configure the operating system in such a way as to minimize unnecessary disk access.

I will show the general principles of what to do and what to try to avoid, and you yourself will set up your system…

I will show the general principles of what to do and what to try to avoid, and you yourself will configure your system to work optimally with a solid state drive.

For example: we know that the Windows operating system actively uses the paging file (the hidden system file “pagefile.sys”) during its operation. What does this mean in relation to the wear and tear of the SSD cells of the drive and everything that we talked about above? And the fact that a separate area of ​​the system flash drive is intensively used (it is often overwritten by some official and unnecessary data and, in fact, is actively worn out)!

What can be done? Right! Transfer the swap file to another (not an SSD drive), as I did, or, with a large amount of RAM, completely refuse it (set to “0”)?

We go further: the defragmentation procedure is not only not needed for this type of device (they have the same access speed for any cell no matter where the final file is located), but it is simply harmful. For the same reason as described above. Excessive (idle) access to the disk only further reduces its limited resource. So – turn off the corresponding defragmentation service. It will also be nice to turn off file indexing, which is needed for faster searches, but do we use it often?

The principle, I think, you have caught. And now I would like to show you a small program “SSD Mini Tweaker” (tweaker-optimizer), which optimizes the operation of an SSD drive in this way. It is enough to put down the necessary checkmarks in front of the corresponding items and click the “Apply changes” button.

The computer will restart and the changes will take effect. The program is remarkable in that it has a Russian interface and detailed help in Russian. So, at any moment you can familiarize yourself with the function that you are going to disable or leave enabled.

You can download the utility from our website. In the archive – versions for 32-bit and 64-bit systems and a help file in Russian.

Since we have devoted so much time to the question of the optimal use of the disk and the wear of its memory cells, I cannot but present you another interesting development. The SSD Life Pro program, the main task of which is to keep track of the disk’s operating time and report the approximate date of its failure.

Unfortunately, the utility was unable to calculate the “lifetime” of my Plextor, but showed some other interesting information.

What do we see here? The entry “FW: 1.00” is the version of the firmware of the disk, the occupied and free space on it, the total operating time from the first start-up and the number of starts are shown below. Also pay attention to the TRIM line (must be active), this indicates that the performance of the SSD drive will be optimal.

Below is a screenshot of the same program, but taken from the site of its developer. It shows that the Intel drive correctly passed its SMART parameters to the utility and, on the basis of them, the utility displayed an extended forecast of its status.

As you can see, the drive’s failure was “assigned” on November 7, 2020 🙂

If we click on the link “How do you think this?” At the top of the program window, then we will go to the developer’s site and be able to see (in Russian) how exactly this calculation is performed?

At the end of this topic, we take into account the recommendations of all respected Intel companies, which say that the ideal working conditions for SSDs of a solid-state drive is less than 75% of its data with a ratio of static (rarely changed) and dynamic (often changed) information – 3 to 1. You should not use the last 10-20% of the disk space, as they are necessary for the correct operation of the “TRIM” command. To work, it needs free space to rearrange the data (just like for the defrag function). The general rule is – the more free space – the faster the device works.

At the moment, the SSD is ideally suited as the system partition on which the operating system and programs are installed and that’s all. Data and all work on them should (if possible) take place on a second (HDD) disk. Solid state drives can also be used efficiently on servers to cache static data.

And now, let’s take a quick look at why the more expensive SSD models of solid-state drives have such excellent speed performance and how are they still different from their “younger” brothers?

Firstly: this is the same intelligent drive controller chip that can be designed as a multi-channel one i.e. – Can write data simultaneously to each flash memory chip of a disk. As a result, the overall performance of the device will be equal to the speed of one memory chip multiplied by the number of controller channels. Well, this is to simplify the situation a bit 🙂


DIY computer assembly
Detailed DIY assembly instructions. Let's start it by installing the processor and motherboard. Installing the processor and motherboard. Before you start assembling your computer yourself, let's look at the figure…


Liquid and air cooling of the computer
Now almost every house has a computer. These electronic machines consume kilowatts from our outlets. Of these, only a small part is spent directly on calculations, and the rest is…


Computer video card and its purpose
A video card is that device with which the image is displayed on the monitor. Those. without a video card, we won’t see any text or images on the screen,…


Computer power supply
In this article we will talk about one of the most important parts of a computer - the power supply, you will find out what important role it plays in…