Different physical storage technologies have been emerging over the years. Each of them has its advantages and disadvantages, and these will clearly define the best uses we can give them. ¿ What is the best hard drive technology? As we can imagine, the answer will depend on our needs. Today we are going to talk about the different current technologies, their pros and cons and which one to choose according to what we want to achieve.
Before we start we can review the main variables that we must take into account when we have to choose the most appropriate storage technology. Obviously, the storage capacity is one of the main ones, but it is not the most important. Depending on the needs that we have, the speed of access to the data -or reading/writing- can be critical. Durability or energy consumption is also important variables to take into account, as well as the acquisition price of each unit.
HDD, or Hard Disk Drive
The HDD technology is already very veteran, but it does not stop having its interest in multiple applications. Although little by little new technologies are starting to have more market share even in the domestic sphere, mechanical hard drives are still an interesting option when it comes to having large storage space at reduced prices.
This type of hard drives have more and more capacity and smaller size and work with disks or turntables on which is operated through a drive arm with reading and write head. To move the discs and operate the arm it is necessary to have an engine, and also an I / O controller and firmware to be able to communicate with the rest of the computer for USB flash drive repair.
The discs are organized into tracks - concentric circles - and the tracks are divided into sectors. Thus, each zone of the disk is defined univocally by an address. To read or write, the arm is placed on the indicated sector in the precise track, and the operation is performed.
The advantages of HDDs are its low price and the high capacities it can offer. The disadvantages, on the other hand, are the time of access to the data -both for reading and writing- and the possible mechanical failures. Being based on magnetism to record and read the data, they can disappear before strong magnetic fields. Other disadvantages of the HDD is that they can suffer fragmentation, which causes vibrations and is noisier than other options. They consume more energy than the SSD, which we will see in the next section.
SSD, or Solid-State Drive
The SSD technology gives us data storage device using non - volatile memory such as Flash memory to store data. Therefore, they do not use plates or magnetic disks like the HDD.
By using electrical impulses to access the data recorded in that non-volatile memory and dispense with the mechanical elements, SSDs are faster. The speed of reading and writing SSDs is up to 10 times higher than in the case of HDDs.
The process of updating data in SSDs is more complex than in the case of HDDs. Without going into too much detail, when the data of a given block is updated, it must first be copied to a different block. Then, the original block is deleted and the data is rewritten with the changes in a new block.
Since there is a finite number of times that a block can be rewritten, there are certain processes in charge of levelling the wear, taking care that each block in the memory is rewritten the same number of times.
The advantages of SSDs over HDDs are obvious. In the first place, its greater speed of access to the data allows improving the general performance of the servers. Although the difference is measured in milliseconds, the perception of the user in what processes will be remarkable.
These discs are more reliable than HDDs because they do not use mechanical parts. In addition, they do not need as much ventilation and do not cause vibrations that can affect the reading and writing of the data.
Finally, the energy consumption of these discs is lower than that of the traditional HDD. This is something to take into account in the long term since the electricity bill can be reduced noticeably in certain environments. At the domestic level, logically, there will be hardly any difference in terms of consumption on the bill, but at the organizational level, given the lower refrigeration needs that we are going to have.
The disadvantage of SSDs is their price, higher than HDD units. This variable, however, is being progressively equated, as SSD technology becomes widespread in the market. In addition, the capacity of SSDs is still behind the largest HDD units.
NVMe, improving the performance of the SSD
Finally, we will mention the NVMe interface, a specification for access to the SSD units connected through the PCI Express bus. Thanks to this interface, it is possible to multiply by six the speed of access to the data -measured in IOPS, or input/output instructions per second- and, therefore, improve the performance of the SSD.
In relation to HDD discs, NVMe improves the reading speed up to 14 times in sequential accesses, and up to 2,000 times in random accesses. This makes it the ideal solution for projects with workloads sensitive to latency, where the speed and volume of operations prevail, such as Big Data applications, Internet of Things (IoT) or high-level databases. performance.
To choose the best hard drive technology for our projects we must take into account, first, the needs in terms of performance and storage capacity. As we have seen, for projects sensitive to latency, the most advisable thing is to have SSD disks with NVMe, since the access speed is higher than the rest of options by several orders of magnitude - in the case of random accesses. On the other hand, the higher cost of the SSD solution can be profitable in many ways, especially if we take into account the lower energy consumption and, above all, the greater reliability of the reading/writing processes.