Hard Disk reading and writing materials, I feel really professional and knowledgeable. Different responses. Be sure to remember. You can talk about things in this way. These things are persuasive and professional.

I saw an introduction today. How the hard disk works and the cause of damage, and finally how to set the system to reduce frequency Read/write damage to the hard disk.
Address: http://blog.csdn.net/uiiang/archive/2004/10/19/143302.aspx
I feel really professional and knowledgeable. Different responses. Be sure to remember. You can talk about things in this way. These things are persuasive and professional.

Today, I just got on the broadband network and finally can download large-capacity software. I usually like watching cartoons. I have never used BT downloading. I asked my classmates when I first got online and went to Mainland China. I heard that BT download will damage the hard disk, so I asked a csdn friend. The following information is provided by a friend on csdn. I will not be entitled to the copyright. In the previous article, I explained the working principle of the hard disk and the cause of the damage. Finally, I explained how to set up the system to reduce the damage.

Why does frequent read/write damage the hard disk?
The lifetime of the head is limited. Frequent reading and writing will accelerate the wear and tear of the head arm and head motor. Frequent reading and writing to a disk area will increase the temperature of the area, it will affect the stability of the magnetic medium in this area and lead to read and write errors. The high temperature will make the area closer to the disc due to hot expansion (under normal circumstances, the head and disk surface are only several microns, coming closer ?), It also affects the Data Reading sensitivity of the thin-film head, changes the clock speed of the crystal oscillator, and causes the failure of the hard disk circuit components.
A large number of tasks may lead to premature damage to the IDE hard disk. Due to the shortcomings of the IDE hard disk, too many task requests will lead to an increase in the seek failure rate and lead to frequent resetting of the head (reset means the head is restored to 0)
Track to retrack) Accelerate head arm and head motor wear.

Let me talk about how modern hard drives work.
The current hard drive, whether it is ide or SCSI, uses the "venturet" technology, has the following features: 1. The head, disc, and motion mechanism are sealed. 2. The flat and high-speed rotating disk slices are flat and smooth. 3. The head moves radial along the disc. 4. The head starts or stops the Contact Type of the disk, but does not directly touch the disk.
Disk: a hard disk is a magnetic particle attached to an aluminum alloy (new materials also use glass) disc on the surface. these magnetic particles are divided into several concentric circles called magnetic channels. On the magnetic channels of each concentric circle, there are as many small magnets in any arrangement. They represent the States of 0 and 1 respectively. When these small magnets are influenced by magnetic forces from the head, the direction of their arrangement changes accordingly. Use the magnetic force of the head to control the specified direction of some small magnets so that each small magnet can be used to store information.
Disk body: the disk body of a hard disk is composed of multiple disks. These disks overlap and are placed in a sealed box. they rotate at a high speed under the drive of the spindle motor, the conversion speed per minute reaches, 5400, or more.
Head: The head of a hard disk is used to read or modify the state of magnetic material on the disk. Generally, each disk has a head, starting from the top and starting from 0. When the head stops working, it is in contact with the disk, but is in a flying state during work. The head is started and stopped by means of a contact type in the landing area of the disc. The landing area does not store any data, and the head is started and stopped in this area. There is no problem of damage to any data. When reading data, the disk is rotated at a high speed. Due to the exquisite aerodynamic design of the head movement, the head is in a "flying state" at a height of 0.2-0.5 microns from the data area of the disk. It does not wear out due to contact with the disk surface, but can read data reliably.
Motor: the motor in the hard disk is a brushless motor, and the mechanical wear is very small under the support of high-speed bearings, can work continuously for a long time. High-speed rotation of the disk has a significant gyro effect, so the hard disk is not suitable for motion, otherwise it will increase the bearing workload. The track-seeking feed motor of the hard disk head mostly uses the audio ring rotation or linear motion stepping motor to precisely track the track of the disc under the adjustment of the Feed Tracking. Therefore, do not have an impact collision when working on the hard disk, be careful when moving.
The principle is mentioned here. Do you understand it?
First, there is no contact between the head and the data zone, so there is no wear problem.
Second, when the hard disk is turned on, the rotation of the spindle motor can reach 4500 or 7200 RPM, which has nothing to do with whether you use flashget or Ed, as long as one power supply, they are switching. their Wear and Tear are also independent of the software.
Thirdly, the movement of the head under the control of the track-finding motor moves back and forth from left to right, and is very small. It starts from the inmost (landing area) of the disc and moves slowly to the outermost layer, then slowly move back, one track goes to another track to find data. There will be no large-scale jumps (not frogs ). Therefore, its wear and tear can also be ignored.
So how does heat come from?
The first is the rotation of the spindle motor and the tracking feed motor. The hard disk temperature is mainly because of this.
Second, the friction between the high-speed rotating disk and the air. This is also the main factor.
And hard disk read/write ???
Sorry, you can ignore the heat value !!!!!!!!!!
The hard disk read operation is the change of the magnetic field on the disk affects the resistance value of the head. In this process, the disk won't heat up, but the head is generated with a little heat due to the current change. What about write operations? In turn, the current strength of the magnetic head is constantly changing, affecting the magnetic field on the disk. Because electromagnetic induction is used in this process, the heat of the magnetic head is high. But the disk itself will not heat, because the permanent magnet on the disk is cold, not because of magnetic field changes and fever.
However, in general, the heat produced by the head is a little cool compared with the previous two.
Heat can be transmitted by radiation, so will high heat damage the permanent magnet on the disk? In fact, the damage is very small. The temperature of the Permanent Magnet degaussing is much higher than the normal temperature of the hard disk. Of course, if your chassis has poor heat dissipation, you can't blame others.

I have to mention several mistakes of someone here:
I. High temperature affects the resistance sensing sensitivity of the head. Therefore, a read/write error occurs, which is irrelevant to the permanent magnet.
II. The so-called thermal expansion will not bring the distance between the disk and the head closer, because the flight of the head is based on the aerodynamic principle, and it is always at a certain distance with the disk under normal circumstances. Of course, if you try to crack the hard disk, then this vibration .....
3. Track seeking refers to moving the hard disk from the initial position to the specified track. The so-called reset action does not happen frequently. Because the physical location of the track is stored in the CMOs, the hard disk does not need to be moved back to the 0 track and then set off again. As long as the head is started, the so-called reset operation is complete. Unless you restart the computer, the reset operation will not happen again.
4. The structure of an IDE hard disk is similar to that of a SCSI hard disk. However, the interface bandwidth of the SCSI hard disk is larger than that of the IDE Hard Disk at the same time, and the scsicard often has something similar to the CPU to reduce the usage of the primary CPU. That's all, so Seagate will use its SCSI hard disk technology on the IDE hard disk.
5. The read/write operations on a hard disk are measured in the sector of the cylindrical disk. The cylinder is the concentric track with the same radius of all the magnetic surfaces in the entire disk, and dividing each track into several zones is the so-called sector. The write operation of the hard disk is to fill one sector first, and then write the next sector of the same cylinder. Before the full write of one cylinder, the head will not be moved to another track. Therefore, the storage of files on hard disks is not stored in a continuous manner as most people think (from the perspective of users, but from the perspective of the underlying operating system, it is not stored continuously ). Therefore, when flash get or Ed runs more threads, The seek of the head is generally not much larger than when you play a game. Of course, this is just a simple download or upload. But in this process, who can ensure that they will not start other software that requires hard disk reading and writing? Maybe many people like to play games or listen to songs while downloading them? Not to mention that Windows itself needs to read and write virtual memory files frequently. Therefore, you can download from FG or Ed, and the hard disk will not be much better than the normal time.
Sat. Why is it difficult to enable too many threads for flash get and why is the hard disk read/write frequency for Ed. First of all, if there are many threads, the CPU usage will be high, and page-changing actions will be frequent, so that the virtual memory reads and writes Will be frequent. As for why, I should know the operating system principles, so I won't talk about it here. What about Ed? Download a file from several people at the same time, and several others are downloading your file at the same time, which is similar to opening multiple threads in FG. So the hard drive lights flash. However, the current hard disk is cached, and the data is not immediately written to the hard disk, but stored in the cache first, and then written to the hard disk after a certain amount. In FG, how can we set it? In fact, it is written to the cache first. However, this process also requires CPU intervention. Therefore, the setting time is too short and the CPU usage is high. Therefore, the hard drive lights are still flashing because the virtual files are being read and written.

7. Hard Disk reading and writing are frequent, and the head arm moves frequently under the drive of the track-seeking servo motor. However, for machinery, there is not much wear loss. Unless your hard disk itself has mechanical faults, such as force arm deformation (the most common faults of water products ). The real wear loss lies in the head. The changing current will lead to its aging, but compared with its lifetime ..... It should also be within a reasonable range. Unless the head hits the disk body due to vibration.
8. The most serious impact of high temperature is the mechanical circuit, especially the circuit board outside the hard disk. The above integrated block will accelerate aging at high temperature. So although a glass hard drive of IBM has a bad track, it will disappear immediately when a software is used. If it is more serious, it will be normal to change the circuit board. This is the reason.

It is too tired to write so many words.
In short, hard disks may affect the service life because of poor environment and improper maintenance, but this is definitely not a software error.
Flashget, Ed, and FTP are also good. Although they read and write hard disks frequently, they do not hurt hard disks much as you normally play games. even more clearly, they can ignore the so-called hard drive loss. don't worry about the flash of hard drive lights. otherwise, the servers that provide Web Services and FTP services have a large hard disk read and write capacity, which is far from playing games. The hard disks of the software are comparable.
The hard disk has a parameter called continuous fault-free time. It refers to the maximum time from the start of the hard disk to the occurrence of a fault, in the unit of hours, MTBF is short for English. Generally, the MTBF of a hard disk is at least 30000 or 40000 hours. For details, refer to the parameter description of the hard disk manufacturer. This continuous fault-free time can be divided by yourself to see how many years it is. Then, let's think about how long your hard disk has been working continuously for the longest time.
Currently, I have been running my machine for one year. Except for several shutdown times and several minutes to clean up the dust, I have never stopped it (using golden to 6 generations and 40 GB ). In addition, there are three servers that use SCSI hard disks, which have not been stopped for two consecutive years. The hard disk's calorific value is not comparable to the ordinary IDE Hard Disk (10 thousand RPM hard disk ).
In this regard, I think I have a say.

Finally, let's add a few points:
I. Hard drive is best not to buy goods or repair. Water and cargo are very unsafe during transportation. Although it seems that there is no damage, it may cause damage to the mechanical body due to various factors during transportation. Not to mention repair. To be honest, those who complain that hard disks are easy to damage should first check whether your hard disks are the same.
II. The hard drive work environment must be clean and tidy, especially when there are frequent power outages and a lot of dust. The chassis should be cleaned up every one or two months.
3. Hard Disks are most afraid of vibration and high temperature. So the environment is better, especially the chassis should be solid, so as to avoid too much earthquake. Do not shake the computer desk.
4. Hard Disk fragments should be sorted frequently. There is a misunderstanding of most people. Most people think that hard disk fragments will increase hard disk wear, but this is not the case. The increase in hard disk fragments only results in the hard disk reading and writing taking less time than the fragmentation. The hard disk loss can be ignored (here I will only say one fact, currently, the most commonly used operating systems for servers on the network are UNIX, but there is no disk fragment software in UNIX. Even Microsoft's NT4 does not exist ). However, due to the frequent movement of the head, the read and write time is increased, so the CPU page change operation is frequent, resulting in virtual files (here, the correct statement is actually a page change file) frequent read/write operations increase the load on the hard drive head. This is the disadvantage of hard disk fragmentation.
5. Avoid sudden power outages, cold boot, and other things that increase the CPU load during hard disk read/write (for example, listening to songs when playing games or playing large 3D games when downloading ), these damage to hard disks is even greater than most people think. I won't talk about the reason, and typing is too tired.

In short, as long as you pay attention to the use of the hard disk, the hard disk will not be so fast to talk to us about Byebye. Of course, if the quality of the hard disk itself is not good, then I have nothing to say.

 

Mitigate hard disk damage caused by Bt downloads:
1. Set a large cache Value

BT multi-threaded download and upload, Windows Default kb cache is obviously not enough. If more than MB exists, you can set the cache to a greater value, which can reduce the read/write frequency of the hard disk. Generally ~ 16 Mb is the best. I set it to 32 MB and enable Four Bt processes at the same time. The hard drive indicator flashing is still normal. The procedure is as follows (taking Windows XP as an example ):

In "Start> Run", enter "Regedit", open "Registry Editor", and open "hkey_local_machi" in turn.

Nesystemcurrentcontrolsetcontrolsession Manager

Memory Management, find "iopagelock-limit" (if not, create "iopagelocklimr" with the DWORD type), change its value to a new cache value, and restart it (note: you must enter a hexadecimal value. 32 MB corresponds to 8000, and 16 MB corresponds to 4000. We strongly recommend that you back up the Registry first ).

For Windows ntsystems, run “regedit.exe in the winnt”directory of the guest system disk to go to the Registry Editor, find the string "minlnitsize", and change the key value from "4" to "20" (hexadecimal value, equivalent to 32 in decimal format) you can.

2. Lower priority

BT occupies a large amount of resources, which may affect the system performance. If you are using Windows 2000/XP, you can use the task manager to reduce the priority of all btsoftware, so that even if it occupies resources, it will not have a big impact on other applications. You can use CTRL + ALT + DEL to start Windows Task Manager. On the process tab, right-click the btsoftware process, select "low" or "below standard" in the "set priority" menu.