How did SSDs surpass HDDs to become the next generation of storage?

H

SSDs are much faster, cooler, and more scalable than hard disks (HDDs), and thanks to technological advances in interfaces, memory, controllers, and more, they’re the next generation of storage. SSDs are likely to continue to improve in performance in the future, as they offer greater efficiency and reliability compared to HDDs.

 

If you’ve been shopping for a computer recently, you’ve probably heard the phrase “make sure you get one with an SSD”. Before SSDs became popular, hard disks (or HDDs) were the dominant method of storing data due to their overwhelming speed and storage capacity. First developed by IBM in 1956, HDDs have evolved steadily over the decades to become the standard for data storage, and many businesses and home users have relied on them to efficiently store large amounts of data. Over time, however, the limitations of the technology began to show and the need for new alternatives began to emerge.
Despite the performance of hard disks, SSDs, which have been commercially available for less than a decade, have become the next generation of storage media with speeds that are dozens of times faster than hard disks. And unlike hard disks, whose performance has stagnated, SSD performance is expected to continue to improve. The invention and adoption of SSDs has revolutionized the IT industry in particular. For example, data centers that adopt SSDs are able to effectively support more customers due to much faster data processing speeds. So how do SSDs have such a high potential for growth?
A solid state drive (SSD) is made up of three main parts: the interface, memory, and controller. The interface is responsible for giving and receiving memory from RAM. Traditional HDDs didn’t need fast interfaces due to their limited internal performance, but the explosive speed of SSDs requires better interfaces. These interfaces include SATA, SATA Express, which is faster than SATA, and M.2, which is often found in devices like laptops due to its small size. These interfaces determine the speed of data transfer between the computer and the SSD, which in turn determines the overall performance of the SSD. The introduction of the Non-Volatile Memory Express (NVMe) protocol in particular has enabled SSDs to realize much faster data transfer rates than previous generations of interfaces.
Memory is where data received from the controller is stored and where data is sent to the controller. Prior to the commercialization of SSDs, some products used RAM-based memory. However, due to the nature of RAM, which loses data when power is lost, SSDs are now only made with flash memory. Flash memory is a non-volatile memory, which has the advantage of retaining data even when the power is turned off. There are three types of memory: Single Level Cell (SLC), Multi Level Cell (MLC), and Triple Level Cell (TLC). Memory is made by attaching hundreds of millions of small units called cells, and in the case of SLC, a single cell can store one bit of information: ‘on’ and ‘off’. In MLC, a cell can store two bits of information, and in TLC, three bits. More recently, a type of memory called Quad-Level Cell (QLC) has emerged that can store even more data. However, the more data you store in a cell, the shorter the cell’s lifespan, and the slower the data read and write speeds tend to be. To address these issues, SSD manufacturers are working on various lifetime extension technologies and performance optimization methods.
The controller is the interface between the memory and the interface. The controllers in SSDs are much more complex than those in HDDs. Due to the finite lifespan of flash memory described earlier, it is essential to prevent only certain cells from being written to over and over again. It also needs to remember the address of the specific data and read it back. Creating an efficient algorithm to name these addresses is difficult, so the performance of the controller is directly related to the performance of the SSD. Recently, researchers have been working on optimizing controllers using AI technology, which could further improve the lifespan and performance of SSDs. Also, the way SSDs store addresses is fundamentally different from HDDs. For this reason, older OSes (pre-Windows Vista) will recognize SSDs as HDDs, resulting in inefficient data operations. Therefore, it’s not recommended to install SSDs in computers with operating systems like Windows XP.
So, what are the advantages of SSDs over HDDs? First, they generate less heat, which is a higher threshold than HDDs. HDDs have a motor in the center that spins a thin metal plate and a head that reads the data on the plate through light, but increasing the number of revolutions per minute can overload the motor and damage the plate, and reducing the wavelength of the head’s light to make the physical size of the data smaller can overload the head’s light source and reduce durability. SSDs also generate a lot of heat when they’re running at 100 percent performance, but this doesn’t happen unless you artificially create that situation, so SSDs have the potential to have higher clock counts and higher density cells. Less heat means a more stable system, which is a big plus for users who need to perform high-performance tasks.
Secondly, the capacity that SSDs can have is also quite high. Commercially available 3.5-inch HDDs can have capacities of up to 20TB. HDDs can be produced in sizes larger than 3.5 inches. However, this means an increase in the size of the metal plates, which reduces the number of revolutions per minute of the HDD motor, reducing performance and making it less practical. SSDs, on the other hand, are getting larger as technology improves. This has given SSDs the advantage of large storage capacities that HDDs once had. In recent years, SSDs with capacities of tens of terabytes (TB) have become commercially available, and SSDs are also increasingly being adopted by large data centers such as cloud storage. This makes SSDs an important storage medium not just for personal computers, but also for enterprise data centers.
The only advantage HDDs currently have is their low cost per capacity ratio. HDDs remain an inexpensive way to store large amounts of data, which is advantageous for large backup systems or large file servers. However, the price of SSDs is constantly dropping, and their cost-effectiveness is improving as technology advances. For the time being, SSDs and HDDs will be used together, but in the future, SSDs will be the storage medium you see more often.

 

About the author

Blogger

Hello! Welcome to Polyglottist. This blog is for anyone who loves Korean culture, whether it's K-pop, Korean movies, dramas, travel, or anything else. Let's explore and enjoy Korean culture together!