Advantages and disadvantages of solid-state hard disk and its purchasing guide

First, the introduction of solid-state hard disk

Solid-state hard disk consists of main control chip and flash memory chip. Simply put, it is a hard disk made of solid-state electronic memory chip array. Its interface specification and definition, function and usage are exactly the same as ordinary hard disk (WwW. PC84 1。 CoM), the appearance and size of its products are exactly the same as ordinary hard disks. The storage unit is responsible for storing data, and the control unit is responsible for reading and writing data. It has the advantages of high speed, durability, earthquake resistance, no noise and light weight. Widely used in military, vehicle, industrial control, video monitoring, network monitoring, network terminals, electric power, medical care, aviation, navigation equipment and other fields.

(1) Advantages of SSD:

First, SSD doesn't need mechanical structure, it is completely semiconductor, and it has no data search time, delay time and disk seek time. It has fast data access speed, and has the data reading ability of more than 400M/s, which can reach more than 500M/s at present.

Second, SSD uses flash memory chips, which are durable, shock-proof and fall-resistant. Even if it collides with hard objects, it can minimize the possibility of data loss.

Thirdly, because there are no mechanical parts and flash chips, SSD has no noise and low power consumption.

Fourthly, it is light, 20-30g lighter than the conventional 1.8 inch hard disk, which makes it possible for portable devices to carry multiple solid-state hard disks. At the same time, because it is completely semiconductor, there is no structural limitation, and it can be designed into special electronic hard disks with different interfaces and shapes according to the actual situation.

(2) Disadvantages of SSD:

First, SSD is expensive. At present, the cost of SSD has dropped significantly. 128G SSD is already at the level of 1000 yuan. However, compared with the mechanical hard disk, the price is still very high ~ As a notebook manufacturer, the upgrade cost is much higher than the actual cost after using the solid-state hard disk as an optional accessory, resulting in a price difference of 1,000 yuan between the notebook equipped with the traditional hard disk and the notebook equipped with the solid-state hard disk.

Second, the storage capacity needs to be improved. Nowadays, the traditional mechanical hard disk has moved to 2TB level with the latest vertical recording technology, while the highest record of solid-state hard disk still stays at several hundred GB (2.5-inch SSD product introduced by PQI). Due to the high cost of flash memory, few manufacturers participate in the research and development of this large-capacity SSD product. Even if there are related products, there is still a long way to go before mass production. At present, the most practical storage capacity of available solid-state drives is only a few hundred GB. But the price is high.

Calculation formula of solid-state hard disk life

Second, clarify the misunderstanding!

1. Why is the speed of SSD not always at the highest speed?

A: At present, most SSD manufacturers will claim that their SSD continuous reading and writing speed exceeds 500MB/s, which is quite impressive compared with the speed of mechanical hard disk1000 MB/s. In fact, almost no program is started and executed continuously. In actual use, only when copying and pasting from different disks will the source disk of data continue to read. That is to say, when a file is copied and pasted from disk D to disk E, disk D is constantly reading and writing.

2. If my hard disk is frequently read and written, will the service life of the solid-state hard disk be insufficient and will it be scrapped soon?

A: On some SSDs, you will see that "the chip is only marked with 65,438+00-65,438+00,000 times of reading and writing". So if I apply it to a database or something, maybe the data I often read and write won't be broken soon, so it's not cost-effective for us to buy a SSD. "Of course, the service life is not like this. If it is not safe, it will not be applied to special fields and harsh environments such as aerospace vehicles!

Solid-state hard disks are basically similar to ordinary mechanical hard disks in principle and structure, such as analog sectors and analog tracks. Inside SSD, the core part is the controller, which is the core of the whole SSD, including many architectures, such as reading and writing algorithms and interface definitions. The main impact on life is the number of reading and writing. In the algorithm definition of SSD, only one modification can be considered as real reading and writing.

Solid-state hard disk flash memory has the problem of limited erasing times, which is also the reason why many people criticize its short life. The complete erasure of flash memory is called 1 P/E ratio, so the life of flash memory is based on P/E ratio. The life of 34nm flash memory chip is about 5000 P/E, and that of 25nm flash memory chip is about 3000 p/e. Does it look shorter? This is true in theory, but with the improvement of SSD firmware algorithm, the new SSD can provide less unnecessary writes. Give another concrete example. A 120G SSD needs to be written into a 120G file to be considered as P/E, and ordinary users praise its normal use. Even if you write 50G every day and complete the P/E ratio once every two days on average, there will be a P/E ratio of 180 times a year. You can calculate for yourself how many years the P/E ratio of 3000 will last. I believe that by that time, solid state drives will have been replaced by other things.

At present, the BladeCenter HS2 1 XM blade server provides solid state hard disk (SSD) based on flash memory technology. Compared with the traditional mechanical hard disk, SSD has the advantages of faster, more reliable, better energy efficiency, less heating and quieter. It can run the operating system and any other applications on the blade server. At the same time, it also shows that the service life is no longer our concern.

With the increase of the number of Flash chips, the service life is also increasing. According to some current applications, the service life of a disk can reach more than 6 years, and the algorithm of the controller is constantly improving, which also improves the service life in disguise on the other hand. I believe the service life will be greatly improved in the future. Thirdly, SSD solid state drive optimization.

1, brush the latest official firmware.

Firmware not only directly affects the performance and stability of SSD, but also affects the life of SSD. Excellent firmware, including advanced algorithms, can reduce the unnecessary writing of SSD, thus reducing the wear of flash memory chips, maintaining performance and prolonging the life of SSD. Therefore, it is very important to update the latest firmware released by the government in time.

Optimization of SSD solid state hard disk

There are generally two methods to update the firmware of SSD: using software update in Windows environment and establishing a startup disk (U disk, CD) update. The software update method adopted by OCZ and other manufacturers, while Crucial Ingrid m4 adopts the latter. The update process is roughly to change the startup sequence of the motherboard BIOS to CD-ROM drive priority or USB flash drive priority, and then enter the boot interface and follow the prompts, which is very simple.

2, open the trim command

SSD will be used more and more slowly, which has a lot to do with the working principle of SSD. SSD is new, and the NAND flash memory in it has been erased in advance, so data can be directly written into the flash memory without completing the data clearing step, and the data writing speed is very fast. As time goes by, the unused storage space in SSD becomes less and less, and in many cases, the data in flash memory must be erased before writing, so its performance will be significantly reduced.

On the Windows 7 system, after starting the Trim command on the SSD that supports the Trim command, the operating system can tell the SSD master that the data block is no longer needed after deleting the file or formatting the file. When some files are deleted or the whole partition is formatted, the operating system sends the Trim instruction and the updated Logincal block address during the operation to the SSD master (including the invalid data address), so as to clear the invalid data in the subsequent garbage collection operation, reduce the write amplification and improve the performance.

In Windows 7, the Trim command is turned on by default. If you want to query the current Trim command status, we can enter the command prompt interface under the administrator's authority and enter "fsutil behavior query disables deletion", and then we will get feedback on the relevant query status. The prompt here is "DisableDeleteNotify = 0", that is, Trim instruction is enabled; The prompt is "DisableDeleteNotify = 1", indicating that the Trim command is not enabled. In addition, it is also important to turn on AHCI mode in the motherboard BIOS. Because the native command queuing feature (NCQ) of AHCI can optimize the order in which users send instructions, thus reducing the mechanical load and improving the performance.

Check the device manager -IDE ATA ATAPI controller. If AHCI is turned on, there will be a prompt behind the controller. If not, it will not be opened.

3. Secure erasure

ATA Secure Erase command can be used to erase all user data on the disk. This command will restore the SSD to factory performance (best performance, minimum write amplification). But the effect is only temporary, because the write amplification will gradually increase back after use and finally return to a stable state. However, after SSD is used for a period of time, the files inside are messy and the performance is degraded. At this time, it is necessary to do a safe erase (the system will be reinstalled anyway).

At present, there are many softwares that can provide ATA secure erase instructions to reset the disk WwW. PC84 1。 And the most famous is HDDErase. However, for SSD, resetting once is equivalent to completing a P/E, and frequent erasure optimization is not recommended. The operation process is roughly to change the startup sequence of the motherboard BIOS to CD-ROM drive priority or USB flash drive priority, then insert the software startup device, enter the boot interface, and follow the prompts.

In addition, the Intel SSD Toolbox is the latest Intel SSD management tool officially launched by Intel, which also includes optimization functions. The principle is similar, but it is more convenient to operate because it is software.

Fourth, SSD purchase.

1, look at the main control chip

At present, the second generation master control of SandForce, which has the highest market share, provides a mature master control scheme. Hard disk manufacturers only need to buy a solution, plus their own PCB design, flash memory matching and firmware algorithm, they can manufacture solid-state hard disks. It is somewhat similar to Google's Android open source model, but the disadvantage is the same, that is, the same main controller is compatible with various chips and firmware, so the performance of hard disk products controlled by SandForce is also uneven. In addition, there are Marvell master control and Intel master control, but there are fewer products, but the performance is quite powerful.

2. Look at the flash particles

In the past, the flash granules used in solid state drives were different from WwW. PC84 1。 CoM, such as 25/34nm process, MLC/SLC, synchronous/asynchronous, ONFI/Toggle mode and so on. The data transfer rate of different flash memory particles is very different. Asynchronous ONFI particles are only 50mt/s (early Intel or Micron particles), synchronous ONFI 2.x particles can reach 133MT/s ~ 200MT/s (Intel or Micron particles), and asynchronous switching DDR 1.0 particles can also reach 133MT/s.

Types, advantages and disadvantages of liquid crystal backlight (LCD, CCFL, LED) (1)

The principle of liquid crystal backlight display The biggest difference between liquid crystal and plasma is that liquid crystal must rely on passive light source, and plasma TV belongs to active luminous display equipment. At present, the mainstream LCD backlight technologies on the market are LED (Light Emitting Diode) and CCFL (Cold Cathode Fluorescence).

Lights) two categories.

Cold cathode fluorescent lamp (cold cathode fluorescent lamp; CCFL)

Traditional liquid crystal displays are all backlit by CCFL (cold cathode cathode fluorescent lamp). CCFL has two main backlight designs: "side-entry" and "direct-down". However, due to the light guide design, the light loss rate is higher, which further limits the backlight brightness. The larger the panel size, the lower the brightness. Only suitable for 8 ~ 15 inch TFTLCD panel, that is, for personal viewing, such as notebook computers and desktops. However, when the LCDTV used for home viewing is large, it is side-entry.

However, the larger the LCD, the higher the proportion of its backlight module, which refers to the direct CCFL backlight module. According to statistics, the direct CCFL backlight module is also used, which only accounts for 23% of the total cost at 15 inch, but it increases to 37% at 30 inch, and it is expected that the backlight module will account for 50% at 57 inch. Therefore, the direct CCFL backlight is only suitable for medium-sized LCD TVs of about 30 inches, and is not suitable for larger-area designs. At the same time, CCFL uses mercury gas discharge to generate illumination. Although the RoHS standard formulated by the European Union is acceptable as long as the dose of "mercury" is lower than the standard, there is no guarantee that the standard may be raised to zero (completely banned) in the future, CCFL will not be used at that time, or CCFL without mercury must be used.

Even if mercury-free CCFL is technically feasible, CCFL is still a kind of gas discharge electronic lighting with a closed lamp tube, which has limited resistance to external forces. Compared with other solid electronic lighting (such as LED), a large impact will break the lamp tube and make the lighting ineffective. In addition, because the direct type does not need a light guide plate, there is no light loss problem, so there is no need for brightness enhancement film, especially brightness enhancement film is a patented technology of a few manufacturers, which is expensive. The direct type can save the light guide plate and brightness enhancement film, which is helpful to reduce the cost.

However, directly dropping CCFL also has its disadvantages. In order to improve the brightness of the picture, the number of lamps must be increased. However, the result of too dense arrangement of light tubes will not be conducive to heat dissipation. As the distance between the left and right phases decreases, the heat dissipation space must be increased from the thickness level. However, the increase in thickness is equivalent to partially offsetting the advantages of LCD TV: thinness.

By the way, when CCFL light pipe is used in large-inch LCD TV, the length of light pipe must also increase with the increase of inches. However, the problems of brightness MURA and color MURA are easy to appear in the middle and both ends of long CCFL light pipe, which will further affect the light uniformity of backlight. In order to keep the uniformity of light continuously, it is necessary to use diffusion film to enhance the uniformity of light, but diffusion film will also bring the loss of transmittance and reduce brightness. The decrease of brightness can only be enhanced by increasing the number of lamps, but as mentioned above, it will be more difficult to design heat dissipation, increase the thickness of backlight module and even increase power consumption. As far as we know, the power consumption of CCFL backlight module has accounted for 90% of the total power consumption of LCD TV. Therefore, changing backlight technology is one of the directions to change LCD image quality at present.

Light emitting diode; (LED); Light emitting diodes)

Because CCFL backlight has many side effects, the industry is also looking for various new backlight technologies. LED is one of the feasible solutions, such as Sony's Qualia series TV, which is a high-end large-size (40-inch, 46-inch) LCD TV with WLED backlight technology. At present, the research and development of LCD with LED backlight technology has reached a substantive stage, and we can already see the related products displayed at CES in 2007.

LED backlight has many advantages. First, solid-state electronic lighting has higher impact resistance than CCFL, so there is no need to worry about the environmental protection regulations of mercury gas and UV ultraviolet leakage. At the same time, it surpasses CCFL in color saturation and service life. In addition, LED

As long as the DC voltage can drive, unlike CCFL, which needs alternating positive and negative voltages, even if only the forward driving voltage is considered, the demand level of LED is lower than CCFL. Furthermore, the brightness of LED only needs pulse width modulation; (PWM); PWM) mode, and the afterimage problem on the TFTLCD display can be suppressed in the same way. However, the brightness level of CCFL is more complex, and the afterimage cannot be suppressed, so it must be suppressed in another way.

Although LED backlight has many advantages, it also has disadvantages. First, luminous efficiency. As far as the same power consumption is concerned, LED is not as good as CCFL, so the heat dissipation problem will be more serious than CCFL. In addition, LED is a point light source, and it is more difficult to control the uniformity of light compared with CCFL line light source. In order to achieve the uniformity of light as much as possible, it is necessary to carefully select the types of LEDs produced and use a large number of LEDs with the same characteristics (wavelength and brightness) for the same purpose. Fortunately, the luminous efficiency of LED is still improving, which can reach above 100ml/W at present, so that the color saturation can be better and the WLED arrangement of backlight can be looser, so that the problems of power consumption and heat dissipation can be alleviated. After the manufacturing yield continues to improve and mature, the cost of LED with consistent brightness characteristics in fermentation will also be reduced.

Just changing the backlight technology may not be enough to trigger the LCD revolution, so let's look at the development of other LCD technologies. Organic light emitting diode (OLED) refers to organic light emitting diode. Organic light emitting diode display technology is different from the traditional liquid crystal display mode. It does not need backlight, and uses very thin organic material coating and glass substrate. These organic materials will emit light when current passes through them. In addition, the organic light-emitting diode display screen can be made lighter and thinner, with a larger viewing angle, and can significantly save electricity. But at present, its life and price are the bottlenecks that limit its development in LCD.

Organic light emitting diode (OLED) is another striking panel application technology, and the realization cycle of small-size panel is earlier. According to the customer's plan, there will be more models coming out in 2008 ~ 2009, but the subpanel is still the main one. Even if the models and shipments have increased significantly compared with the present, the market share will not exceed 10%. Originally, organic light-emitting diodes are better than TFT- LCD because of its thinness, contrast, viewing angle and power saving. Concerned by the industry, it is thought that it will replace TFT-LCD, and it was also put into research and development in early years. However, on the one hand, the technology of organic light-emitting diodes itself has encountered bottlenecks, and life problems need to be overcome; On the other hand, the continuous progress of TFT-LCD technology can now provide excellent contrast and perspective, resulting in the demand for organic light-emitting diodes has never been greatly improved, and the market is small and the supply exceeds demand, which is limited by price competition; The original investment operators can't escape the fate of dissolution and downsizing. In the past, Taiwan Province Shenghua Technology invested in the establishment of Shengyuan to invest in the research and development of organic light-emitting diodes. See organic light-emitting diodes and TFT-LCD can not be compared, especially the cost gap is large. In terms of specifications, TFT-LCD can easily achieve a viewing angle of 170 degrees, a contrast ratio of 500: 1, and increase brightness. Can also be made very thin, the response speed is slightly slower, but it can reach the acceptable range of human eyes. Therefore, Sheng Yuan was also closed, leaving only a few R&D personnel to return to Shenghua to develop materials. If the life and price of organic light-emitting diodes can be greatly improved in the future, there is still a chance; At present, it is limited to products with particularity and emphasis on novelty; I haven't seen it for a long time.

And amoled (Active Matrix/Organic Light Emitting Diode) Active Matrix Organic Light Emitting Diode Panel (AMOLED) is called the next generation display technology, including Samsung Electronics, Samsung SDI and LG Philips. At present, apart from Samsung Electronics and LG Philips, which mainly focus on developing large-size AMOLED products, both Samsung SDI and AUO focus on developing small and medium-sized products. Judging from the current performance of finished products, if the cost of AMOLED can be effectively controlled, the traditional LCD panel technology will be greatly challenged.