Hard disk, also called storage device. It is an indispensable thing for digital products in today’s era, cell phones, computers, cameras, servers, wherever you need to store data, the existence of hard disk is indispensable. Hard disk can be divided into two kinds of mechanical hard disk and solid state hard disk, in the past a long time, mechanical hard disk has been a computer is indispensable, in recent years, with the breakthrough of solid state hard disk technology, more and more people choose to use the solid state hard disk loaded with mechanical hard disk is only used as an auxiliary storage device.
Why do all laptops use solid state drives? What are the advantages of solid state drives over mechanical hard drives? What experience will these places affect us? In this issue, we will talk about mechanical and solid state that point.
First, the history of hard disk development
The beginning of hard disk history, naturally belongs to the mechanical hard disk, as early as 1956, the world’s first hard disk IBM RAMAC 350 was born.
Figure 1. IBM RAMAC 350 (source network)
Compared with the modern hard disk, this thing is absolutely considered a behemoth, as big as two refrigerators, weighing more than a ton, but the capacity is only 5MB size. It marked the beginning of the hard disk era, when computers abandoned the primitive storage tools of perforated paper and magnetic tape, and in the 1980s, hard disks were about the same size as they are today.
Figure 2. Mechanical hard disk in the 80s (source)
In 1988, Greenberger and Fell discovered the giant magnetoresistive effect, a technology that enabled mechanical hard drives to develop by leaps and bounds, with great increases in speed and capacity, and the rapid popularization of mechanical hard drives. At the same time, the germ of the solid state hard disk also produced, but this time the solid state is stacked with memory, until the eve of the 21st century, only some manufacturers began to use flash memory instead of memory to manufacture solid state hard disk, flash memory compared to memory can be done after the power outage of data is not lost.
After 2011, there was a natural disaster in Thailand, the largest manufacturer of mechanical hard disk parts, which caused prices to skyrocket, which allowed the SSD industry to take off.
Figure 3. SSDs produced by Samsung (Source: Internet)
II. Mechanical Hard Disk Working Principle
The internal structure of a mechanical hard disk is mainly composed of a motor, a disk, a head arm, and a head.
Figure 4. Internal structure of mechanical hard disk (Source network)
When a mechanical hard disk is in operation, the magnetic head is suspended a few nanometers above the disk surface. The disk surface is divided into numerous small compartments, which are filled with small magnetic particles. These magnetic grains on the disk have a certain polarity, which is noted as 0 when the grains are polarized downward and 1 when the grains are polarized upward.
Figure 5. Magnetic head read or write schematic (source web)
When reading data, the read head reads the data by reading the polarity of the magnetic grains. The write head can use the magnetic field generated by the coil to change the polarity of the grains so that it can write or rewrite the data.
The disk itself is divided into a myriad of sectors and tracks in order to pinpoint the location of data on the disk surface. Let’s say that the data is stored on sector seven of the fifth track of the disk. Then the magnetic head would first swing over the fifth track and wait for the seventh sector to turn over. It is only when the seventh sector turns under the head that the data can be read.
The mechanical structure of a mechanical hard disk brings one fatal disadvantage in the first place – the fear of wobbling. Because the magnetic head is infinitely close to the surface of the disk during high-speed reading and writing, shaking can cause read and write data to be lost, so mechanical is very unfriendly to portability. In addition, the mainstream mechanical hard disk rotation speed is 7200 rpm, although this figure seems very fast, but this speed for a few thousand MHZ memory and a few GHZ CPU is too slow, so the mechanical hard disk inevitably becomes the shortest board to slow down the system running speed.
Third, the working principle of solid state hard disk
These shortcomings of the mechanical hard disk, which happens to be the advantages of the solid state hard disk. Solid state hard disk is actually a circuit board, so manufacturers can make it into any shape. Take the m.2 interface, it can be divided into these specifications.
Solid state hard disk has no mechanical structure, no noise, no fear of vibration, high portability, and the entire electronic interaction of data. The storage unit of a solid state drive is a floating gate transistor, whose internal structure is shown in Figure 8: a floating gate layer for storing electrons, a tunneling layer for restricting the free movement of electrons, an insulating layer for isolating electrons, a control pole G, a substrate P, a source D and a drain S. When storing data, the floating gate layer is used to store the data in a floating gate.
The number of electrons in the floating gate layer above a certain value is counted as 0 and below a certain value is counted as 1 when storing data.
Write Data
To write data, a high voltage (e.g., 20 V) is applied to the control pole G, so that the electrons in the substrate are attracted to pass through the tunneling layer and enter the floating gate layer, after which the electrons are hindered by the insulating layer and can only be imprisoned in the floating gate layer. The existence of the tunneling layer ensures that the imprisoned electrons will remain in the floating gate layer even if they lose the attraction of the high voltage. This completes the process of writing data to a cell.
Erasing Data
The process of erasing data is to release the trapped electrons from the floating gate layer. Simply by applying a high voltage to the substrate, the electrons are attracted by the electric field and return to the substrate through the tunneling layer, thus returning the cell to its original state.
It is because of this read and write principle that many people worry about the longevity of SSDs. This is because when writing data, SSDs do not directly overwrite the original data, but erase it first and then write it. Every time a cell goes through an erase-write cycle, the floating gate layer will inevitably leave some charge behind. Just like writing on paper with a pencil, writing and erasing and writing, the paper will eventually rub through. The same is true for solid state drives.
However, this is for our ordinary users do not have to worry about. Take a 512G solid state:
Daily write volume 140G also need ten years to exhaust the life, besides, many users do not have such a high write volume.
So SSDs can still be used with confidence.
Reading data
When there are only a few electrons in the floating gate layer (storage state 1), we apply a low voltage at the control pole, which causes the electrons to be attracted to the underside of the tunneling layer, but due to the low voltage the electrons are unable to pass through the tunneling layer, thus forming a conductive trench between the source and drain, and a current can be formed by applying a voltage between the source and drain.
And when a large number of electrons are stored in the floating gate layer (storage state 0), the same low voltage is applied at the control pole. Since the electrons stored in the floating gate layer have a repulsive effect on the electrons in the substrate, a conductive channel cannot be formed in the substrate, preventing the generation of current between the source and drain.
Therefore, if a current is detected during reading, the cell state is 1 and vice versa 0. This completes the reading of the data in the cell.
Data Writing in the Storage Module
Since the number of storage units in the SSD is very large, if each module is set up with an independently controlled switch, this will inevitably lead to a redundant module structure and reduced storage efficiency. So the rewriting of data within the storage module is controlled by the module.
A schematic diagram of a storage module is shown in Fig. Each row is a storage byte and contains a total of eight storage units. Each column of storage units connected in series structure, the top row and the following row of transistors, used to control each bit (column) of the write operation; each row of storage units of the gate by an electrode unified control.
When you want to perform a write operation on a particular cell (e.g., the first column of the fourth row), you give the control transistor of the first column a zero bias so that a high voltage is applied under the gate of the fourth row so that the floating gate of that cell can store charge. While the other control transistors are given a low voltage of 2V so that a current is formed and due to channel effect the other cells in the fourth row cannot store charge.
IV. Summary
SSDs interact electronically throughout, and electronic signals are much faster than mechanical structures like head arms and disks. If the data is randomly dispersed in all corners of the disk, then the mechanical hard disk needs to go through a number of seek and address, many times waiting for the sector to rotate to the bottom of the head, so the mechanical hard disk in the read dispersed files, the performance is very weak and slow, that is, random read and write performance is low.
In daily life, the most impact on our experience is the random read and write capability of the hard disk, like many software is composed of many small files, the most prominent of which is our windows operating system, open the computer c disk can be seen, the operating system is composed of countless dispersed small files, these files are not big but very dispersed, if it is a mechanical hard disk to read these dispersed data, as each files in different locations on the disk, limited by the mechanical structure of the disk, reading these will produce a long wait. This is why the system disk must be a solid state drive.
In addition, with the innovation of solid state technology, the price is getting lower and lower, the capacity is getting bigger and bigger, and ordinary users don’t need to worry about the service life of the solid state, these advantages make the solid state hard disk will be the first choice of users.
原创文章,作者:大柱,如若转载,请注明出处:What are the advantages of solid state drives? [Science]