The storage of the databased takes place in the files. Moreover, we have file formats for their proper storage. These file formats contain the records. If we talk about the physical storage of the databases, then it happens in the electromagnetic format available on the same device. There are three categories under this broad area of storage of the databases. Read the concept of normalization in database management system to get an understating of the system.
These are the primary memory, secondary memory, and the tertiary memory. We will discuss them in the article. Read on and get to know the concept.
1. Primary Storage
The type of primary storage which gets directly accessed by the CPU is the prime storage. The registers are present in the internal memory of the CPU. The other elements which make the central processing unit are fast memory or the cache, and the main memory, i.e. RAM is also present in it. The CPU can directly use and reach these elements. It is because of their presence on the chipset CPU and the motherboard. There are several features of this type of storage. These are volatility, its ultra-high-speed. However, the size of the memory is small. For the uninterrupted use and access, we have to give it a continuous power supply for the maintenance of the state. We lose all the data if we experience a power failure.
2. Secondary Storage
For the storage of data to use in future, we use the secondary storage. We can have the backup of the data by storing the data in the secondary storage devices. These are the optical disks, magnetic disks, flash drives, magnetic tapes, and many more. These are the devices which are not contained in the CPU chipset, motherboard.
3. Tertiary Storage
For the storage of data in huge volumes, the use of the tertiary devices is the best. They are the devices which are used from the outside of the computer system. This is the main reason of the drop of their processing speed. The best use of the system is to keep the backup of the data as well as the entire system. The examples, in this case, are the magnetic tapes and the optical disks.
Hierarchy of Memory
The system computer and its devices come up with the memory hierarchy. The device or the element which have the direct access to the main memory is the CPU. In addition to this, the in-built registers can also be directly accessed by the CPU of the computer system. The CPU speed is much more than the accessing time of the main memory. For the proper functioning of the computer system, we need to minimize the mismatch of the speed. To overcome this, the system is introduced with the cache memory. It is because it is the cache memory which accesses time fast. In addition to this, the cache memory consists of data which the CPU accesses frequently.
It is obvious that the memory which has the fastest access to the CPU comes after spending a big amount of money. The less expensive storage devices are large, and they have the slow speed to offer. In comparison to the registers in CPU and the cache memory, we can store huge volumes of data.
The most secondary storage devices in the list of the memory choices available are the magnetic disks. The name has been given to them because of their magnetization features. They use this concept for the storage of the information with its magnetizable material. We have a spindle in the device, and these are the devices which are placed on them vertically. There is a head which moves in between the disks. It reads and writes the information. It is a thing which is used for magnetizing as well as de-magnetizing the spot which is under its influence. For recognizing the magnetized spot, we use two values. These are 0 and 1.
For the effective storage, we have a well-defined order for the formation of the hard disks. There are several concentric circles over the hard disk plate. The term used to describe those circles is tracks. There are several sectors which make up the tracks and form them. The default memory space in which the sector on the hard disk stores is 512 data bytes.
The full form of the word is Redundant Array of Independent Disks. It is a technology for the connection of multiple secondary storage devices, and they are to be used in the form of a single storage media.
What is constituted in RAID?
We have an array of disks in RAID. Here, the array has several disks in connection to each other. They have to achieve diverse goals in the system. There are several levels of the RAID which are used for the arrays in the disk.
1. RAID 0-
We implement the striped disk array in this level of the technology. For the process to take place, the data is broken down into several small blocks. After this, the blocks are distributed among the disks. The block of data is given to every disk, and the main purpose of the division is to read and write the data in parallel. For the enhancement of the speed as well as the performance by the storage device, we use this level of technology. However, the negative points are also there. We don’t have any parity or the backup at RAID 0.
2. RAID 1-
It is the technique which uses the concept of mirroring. After we send the data to the controller of the RAID, the array gets the copied data. The other name due to the functionality and the feature is mirroring. In the case of the failure of the work by the technology level, it provides pure redundancy.
3. RAID 2-
Under this level of technology, we have the Hamming Distance technique for the records in the Error Correction Code. The data is striped on the various and diverse disks. As we have studied in the zero level of technology, the recording of the data bit in a word on the separate disk happens in the RAID2 also. On the set of the different disks, the error correction codes containing the words of the data are stored. It is commercially unavailable because of its high cost as well as the complex structure.
4. RAID 3-
Under this level of technology, the data is striped onto the different disks. Due to this, a parity bit is generated. After the execution of the level of technology, different disks store the data word of the parity bit. This is the best technique if we want to overcome the failures of the single disk.
5. RAID 4-
RAID 4 writes the entire block on the disks of the data. After this, the system experiences a generation of parity which is stored on the disk which has not been used. Under this level, if technology, we have the block-level stripping and in the previous technology level, we defined the process by the byte-level striping. Remember that both these levels require a minimum of three disks for the RAID implementation.
6. RAID 5-
As the functioning of the previous level of the RAID technology, this level also writes the entire block of data onto the different disks. However, the change in both these levels is that the previous technology level stores the data on the dedicated but the different disks. In this case, the generated bit of parity for the block of striped data is distributed among the data disks.
7. RAID 6-
The extension of RAID 5 is RAID 6. Under this level, we have the generation of the parities. These are then stored and distributed among multiple disks. If we have two parities in the system, then it creates an additional fault tolerance. For the implementation of this level, there is a requirement of at least four disk drives.