What is a Solid-State Drive?

Ⅰ. Introduction and development of SSD

1. Introduction

A solid-state drive(SSD) is a hard drive made from an array of solid-state electronic memory chips. SSD is composed of control unit and storage unit (FLASH chip, DRAM chip). The specifications and definitions of interfaces, functions and usage methods of solid-state hard disks are exactly the same as those of ordinary hard disks, and the shape and size of the products are also completely consistent with ordinary hard disks. It is widely used in military, vehicle, industrial control, video surveillance, network monitoring, network terminals, electric power, medical, aviation, navigation equipment and other fields.

A solid-state drive is a newer, more advanced form of hard drive that can be used as a storage device in a computer. With no moving parts, it accesses and transfers data faster than traditional drives.

2. Development

In 1970, StorageTek Corporation (Sun StorageTek) developed the first solid-state hard drive.

In 1984, Toshiba invented flash memory.

In 1989, the world's first solid state drive appeared.

In March 2006, Samsung was the first to release a 32GB SSD notebook computer.

In 2007, SanDisk released 1.8-inch 32GB SSD products and 2.5-inch 32GB models.

In 2009, a large number of manufacturers poured into the solid-state hard disk market, and solid-state hard disk technology has developed rapidly.

On July 21, 2022, Samsung Electronics announced that the company has successfully developed the second-generation smart solid-state drive (SmartSSD), which will be used to seize the future market in the future.

Ⅱ. The basic structure of SSD

Flash-based SSD are the main category of SSD, and their internals are quite simple. The main body of a solid state drive is actually a PCB board, and the most basic accessories on this PCB board are a control chip, a cache chip (some low-end hard drives do not have a cache chip) and a flash memory chip for storing data. Among these three components, the flash memory chips is the most expensive, and the main control chip is the one with the best technical content and core technology.

1. Main control chip

The main control chip is equivalent to the CPU of a computer, which manages the solid-state hard disk through firmware, so the performance of the main control directly affects the overall performance of the solid-state hard disk.

(1) Function

① Scheduling the load of data on each flash memory chip, so that all flash memory particles can work normally under a certain load, and coordinate and maintain the cooperation of different block particles.

② It undertakes the entire data transfer, connecting the flash memory chip and the external SATA interface.

③ It starts and runs the firmware algorithm, and urges the firmware to complete various internal instructions.

(2) Brand

The technical threshold of the main control chip is relatively high, so there are relatively few main control brands. At present, mainstream main control brands include Silicon Motion, Phison, Marvell, Samsung, Realtek, Intel, and Toshiba (Kioxia).

① Silicon: Silicon Motion is a Taiwanese master control company. Its main control cost is low, and it is welcomed by many domestic SSD. Its manufacturers includes Plextor, Colorful, GALAX, Taipower, Guangwei, Maxsun, etc. It is mainly characterized by good performance and low price, and it is relatively balanced overall.

② Phison: It is also from Taiwan. Starting from the master control design of the USB flash drive to the master control of the solid-state hard disk, the master control chip designed by Phison is famous for its good cost performance.

③ Marvell: Marvell belongs to the high-end series and is now used in Plextor, SanDisk, and Crucial SSDs. The technical strength is strong, the main control quality is stable, but the price is relatively expensive. Consumers who have used Marvell's master control praise its performance and stability.

④ Samsung: As the world's top semiconductor company with a complete ecological chain of semiconductor design, processing, manufacturing, production, and sales, Samsung occupies a major position in the field of solid-state hard disk master chips, and creatively develops products based on Samsung's own A series of master chips for firmware.

⑤ Realtek: It is a newly promoted master control brand in Taiwan. Some of the colorful solid-state drives use Realtek master control solutions.

⑥ Intel: Intel has always been the best choice for "stable and reliable" in everyone's mind. Its high reliability is the biggest feature. The current mainstream models are the 545S and 760P series, both of which offer a five-year warranty.

⑦ Toshiba: Toshiba Storage Corporation changed its name to Kioxia from October. According to analysis, due to the drag on the nuclear power business, Toshiba had to spin off its most competitive flash memory business into Toshiba Storage Corporation and sell nearly half of its shares to Bain Capital of the United States. Fortunately, it just changed a different brand, it still maintains the original quality.

2. Cache chip

Next to the main control chip is a cache chip. Like traditional hard drives, SSD require high-speed cache chips to assist the main control chip in data processing. It should be noted here that some cheap solid-state hard disk solutions omit this cache chip in order to save costs, which will have a certain impact on the performance during use, especially the read and write performance and service life of small files. There are two main types of cache: DDR and Slc-cache.

① DDR: The traditional DDR cache, its speed is almost as fast as the memory stick. If the amount of writing to the hard disk is too large at a certain time, it can be used as a temporary storage medium when it is busy.

② Slc-cache: SLC cache. Because most of them are TLC hard drives, SLC cache is preferred to simulate SLC particles for work. Therefore, when transferring large files, there will be obvious speed drops.

3. Flash chip

Except for the main control chip and the cache chip, most of the other positions on the PCB are NAND Flash memory chips. The classification of solid-state drives has two directions, one is to distinguish from its quality (white, black, original), and the other is to distinguish from technology (SLC\MLC\TLC and other particles).

(1) Quality classification

Flash memory particles NAND/3D NAND FLASH in SSD, there is a simple process, production wafer - wafer test - wafer cutting - Good die packaging - testing - processing, and black chip, white chip, original chip, is gradually in this process.

① Original chip: that is, the NAND chip produced and packaged by the original factory itself. At present, there are roughly Intel, Micron, Samsung, Hynix, Toshiba and other brands that can produce and package NAND chips by themselves. They are generally used in high-end hard drives and server storage, and they are also very expensive.

② White chip: Due to production capacity, the original factory will also sell wafer directly to the brand owner, and then the brand owner will find a packaging factory to package it. In this case, the logo is usually not marked on the NAND chip, which is the so-called white chip. It is a defective product left after cutting the original film. The advantage is the favorable price, and the disadvantage is the short life.

③ Black chip: NAND chip completely packaged by ink Die. It is equivalent to waste products, the price is very low, the life span is relatively short, and the stored data is very easy to lose.

(2) Technical classification

NAND Flash memory chip is divided into four specifications, such as SLC (Single-Level Cell), MLC (Multi-Level Cell), TLC (Trinary-Level Cell), QLC (Quad-Level Cell). There is also an EMLC (Enterprise Multi-Level Cell) that is an "enhanced" version of MLC NAND flash memory, which bridges the performance and durability gap between SLC and MLC to a certain extent.

① SLC (Single-Level Cell), that is, 1bit/cell, uses positive and negative charges, and a floating gate stores 1 bit of information, and the voltage change range is small when writing data. About 100,000 erasing and writing life. The speed is fast, the life is long, and the price is expensive (about 3 times the price of MLC). It mostly used in enterprise-level high-end products.

② MLC (Multi-Level Cell), 2bit/cell, uses a double-layer electronic structure with different constructions of high and low voltages, and has a lifespan of about 3000-5000 times of erasing and writing. Its speed, lifespan and price is average. It is mostly used in household high-end products.

③ TLC (Trinary-Level Cell) is an extension of MLC flash memory, and TLC reaches 3bit/cell. Due to the high storage density, the capacity is theoretically 1.5 times that of MLC, and the theoretical erasing times range from 1000-3000 times. Its speed is slow, its lifespan is short, and its price is cheap. It is currently the mainstream flash memory particle on the market.

④ QLC (Quad-Level Cell): QLC flash memory particles have a higher storage density than TLC, and the theoretical number of erasing and writing is only 150 times. It has a large capacity, a short lifespan, slow speed, and favorable price.

Ⅲ. The lifespan of SSD

Like ordinary hard drives, solid-state drives also have the problem of lifespan. Although most brands provide a 5-year warranty to match the average service life of the product, the way of use is the key to determine whether the solid-state drive can operate for a long time. Generally speaking, the product will be marked with a Terabytes Written( TBW) on the specification sheet, and the number represents the upper limit of the amount of data that each hard disk can read and write.

Therefore, when the TBW is marked as 36.5TB, just divide it by the number of days in 5 years, and you can calculate that the hard disk can read and write 20GB on average every day. As long as you use more than this number every day, it will reduce the overall lifespan. If you still can't know the exact remaining life of the hard disk, you can also use free software such as "CrystalDiskInfo" to check the health of the hard disk to monitor, so as to avoid file loss due to hard disk damage.

Ⅳ. Difference between SSD and HDD

1. Advantages

(1) Read and write speed is faster than HDD

A significant advantage of solid-state drives over hard drives is speed. SSD is 25 to 100 times faster than hard drives. Note that as an HDD moves its heads across the spinning metal platters, it takes extra time to read or write data. This mechanism also fragments the arrangement of the data. Still, due to the absence of mechanical parts and fragmentation, SSD is inherently faster, which means faster boot times, faster file transfers, and better bandwidth.

(2) More compact than HDD

The small form factor is also an advantage of SSD. HDD is naturally bulky due to the magnetic heads and metal platters housed in the housing. SSD, on the other hand, consist of small integrated circuits. The compactness of SSD makes them suitable for ultraportable notebooks as well as consumer electronic devices such as smartphones and tablets.

(3) Stronger than HDD

Long life is another advantage of SSD. Internal hardware damage and possible data loss due to drops and shocks has been observed in hard drives due to mechanical components. SSD is more resistant to drops and shocks, and since they have no moving parts, they are less prone to data loss due to external trauma. This also makes SSD ideal for laptops and other mobile devices.

(4) Noiseless operation

Hard drives produce distracting sounds and vibrations as their read-write heads move and metal platters spin when they are in operation. This is especially true when dealing with large amounts of data. On the other hand, there are no moving parts and no mechanical work in an SSD, making it largely silent in operation.

(5) Energy efficiency

SSD consumes less power than HDD. Remember, HDD has moving parts. That means it needs an extra electrical input to get the mechanical work done and start running. SSD isn’t. Energy efficiency is one of the reasons why SSD is ideal storage media for computers and other mobile devices, where battery capacity and battery life are key technical specifications.

2. Disadvantages

(1) The basic storage capacity is small

The base storage capacity of a hard drive is now 500GB, and manufacturers are pushing to standardize on higher capacities, to 1TB or more. SSD, on the other hand, come in base storage capacities of 64GB and 128GB. This is a significant limitation of SSD. Note that the limited base capacity is why entry-level computers with SSD has less storage available. Although there is SSD in capacities of 4TB and above, they are either rarer than HDD or very expensive.

(2) More expensive than HDD

One of the main downsides of SS is price. On a dollar per gigabit basis, SSD is more expensive than HDD. Specifically, an SSD with similar storage capacity as an HDD can cost twice as much as an HDD. Computers and other devices with SSD are usually more expensive. This is due to the increased manufacturing cost of purchasing SSD.

(3) Concerns about the PE cycle

Another significant disadvantage of SSD is the assumed limit on the number of times data can be written and rewritten. Unlike HDD, SSD cannot store a single bit of data in a particular block without clearing and rewriting the entire block. However, the number of writes to flash memory is limited. Consumer-grade SSD has 3,000 to 5,000 read-rewrite or program-erase cycles, while high-end and more expensive models have 100,000 program-erase cycles. Be aware that once the limit is reached, the integrity of the SSD will deteriorate and data loss may occur.

(4) Availability is lower than HDD

Buying hard drives is easier because they are available at most computer and consumer electronics retailers. Furthermore, the HDD market is flooded by a large number of manufacturers. SSD is not only more expensive, but also harder to find, especially those with higher storage capacity models. High manufacturing costs make SSD unattractive for production, while high end-user price points make them unattractive for retail. However, the market is now leaning towards mobility, and SSD is becoming more available, but not as plentiful as HDD.

(5) Lifespan limit

SSD flash memory has the problem of erasing and writing times limit, which is why many people criticize its short life. A flash memory is completely erased and written once called 1 P/E, so the life of the flash memory is based on P/E. The lifetime of a 34nm flash chip is about 5,000 P/Es, while the lifetime of a 25nm flash memory chip is about 3,000 P/Es. With the improvement of the SSD firmware algorithm, the new SSD can provide less unnecessary write volume. For a 120G solid-state hard drive, only a 120G file needs to be written to be counted as a P/E. Ordinary users use it normally, even if they write 50G every day and complete a P/E every 2 days on average, 3000 P/E can be used for 20 years. By then, the SSD will have been replaced by more advanced equipment.

In addition, although each sector of a solid-state hard disk can be repeatedly erased and written 100,000 times (SLC), some applications, such as LOG records of the operating system, may repeatedly read and write a certain sector. In this case, the actual life of the SSD has not been tested. However, through the management of the storage unit by the balancing algorithm, its life expectancy will be extended. SLC has a write life of 100,000 times, and MLC with a lower cost has a write life of only 10,000 times. The cheap TLC flash memory is only 1000-2000 times. In addition, the multi-level storage solid state that uses full-disk analog SLC to increase the writing speed will face the problem of write amplification, which will further shorten the life span.

Ⅴ. Usage and maintenance of SSD

1. Damage and failure of SSD

When we use the solid-state hard drive in our daily life, we can regularly judge whether the solid-state hard drive is working normally through the following methods:

(1) Whether the built-in disk manager of the computer system can recognize the solid-state hard disk, and whether the capacity, hard disk mode and other parameters are correct;

(2) Whether the SSD can be read and written;

(3) Whether the BIOS can accurately identify the SSD, and whether the capacity and other parameters are correct;

(4) Whether the SSD can be initialized, partitioned, and formatted; if any of these operations cannot be performed, it can prove that the SSD is damaged.

2. A few tips about the usage and maintenance of SSD

(1) In terms of usage:

① Avoid excessive temperature: High temperature may affect the performance and lifespan of solid-state memory. We should try to avoid exposing solid-state memory to high temperature environment, especially during long-term high-load operation.

② Back up data regularly: Although solid-state storage is more reliable than traditional hard disks, it is still recommended to back up important data regularly to prevent accidental data loss.

③ Average use of storage space: In order to balance the erasing and writing operations of solid-state memory, we should try to avoid filling its storage space to the limit. Maintaining a certain percentage of free space helps prolong the useful life of solid-state storage.

④ Avoid frequent write operations: One of the main limitations of solid-state memory is its write life, that is, the number of times it can be erased and written. We try to avoid heavy and frequent write operations, especially with large files or sustained writes. For the average user, this is usually not a big problem, because the write endurance of modern solid-state memory is already quite high.

(2) In terms of maintenance:

① Protection from physical damage: Solid-state memory does not have mechanical parts like traditional hard disks, but it still needs to be protected from physical damage, such as dropping, crushing, etc.

② Update firmware: Some solid-state storage may improve performance and stability by updating firmware. Check the manufacturer's website regularly to make sure the firmware is up to date.

③ Solid-state storage does not require defragmentation: Unlike traditional mechanical hard drives, solid-state storage is not affected by fragmentation and does not require regular defragmentation.

④ Do not need to format too frequently: Solid-state storage does not need to be formatted as frequently as traditional hard disks. Excessive formatting may reduce the life of solid-state storage.

To sum up, although the lifespan of solid-state drives is very long, their use and maintenance are also very important. For users, the most important thing is the correct use and maintenance of SSD.