3.1.2 Secondary Storage Flashcards
4 features of secondary storage
Secondary storage stores data that has to be loaded into RAM before the CPU can access it.
It is always non-volatile.
It typically has a very high capacity compared to memory / RAM, i.e. it can store a lot of data.
Whereas primary storage is integrated into the circuitry of a motherboard, secondary storage is separate from the motherboard.
secondary storage
non volatile long term persistent storage for programs and data
why do we need secondary storage
Computers use primary memory such as random access memory (RAM) and cache to hold data that is being processed. However, this type of memory is volatile, which means it loses its contents when the computer is switched off. General purpose computers, such as personal computers and tablets, need to be able to store programs and data for later use.
SS is non-volatile and can store programs and data when the power is switched off.
ROM is not volatile, but it is read-only, and we often want to change or add to our stored data.
SS allows read and write access, so we can edit and add to our files.
Both RAM and ROM are expensive per GB and it would be too expensive for them to store all of a modern user’s files.
Some forms of secondary storage are relatively cheap per GB so they come in much larger sizes than (say) RAM (e.g. 1 TB HDD v 8 GB RAM). Secondary storage typically has much greater capacity than primary storage.
without SS…
Software would need to be installed each time we use it.
You couldn’t save any files, data, images, videos etc.
You could not back up your user settings (e.g. favourites, your application data).
do all computers need SS
no!
Many embedded systems, for example, have no need to store data in secondary storage. Their instructions can be stored on a ROM chip, and there may be no need to record long-term any data once the device is in use (e.g. a computer controlling a dishwasher).
3 main types of SS and egs
Magnetic storage, e.g. a hard disk drive OR tape drive.
- stores data through the polarity / state of the magnetic layer on a disk
Optical storage, e.g. a CD or DVD drive, blu-ray
- stores data by examining how the surface of a disc affects the light emitted by a laser
Solid state, e.g. a solid state drive or USB memory stick.
- stores data as electric charges
how does an HDD work (magnetic storage)? (what is a platter!!)
Read/Write heads on a moveable arm, for each disk, which move to particular tracks/sectors on the disk.
Multiple magnetised disks. These spin up to 250 times per second (over 100 km per hour!).
A HDD may have one or more platters.
The platter is a metal disk with a thin magnetic coating on each side.
Data is stored in many tiny dots on the coating.
The magnetic state / polarity of the coating can be changed to represent a 0 or 1.
The magnetic state / polarity can be read / changed by the read/write head that moves over the spinning disk.
difference between disk and disc
For magnetic storage (e.g. HDD):
disk
For optical storage (e.g. CD, CD-ROM, Blu-Ray, DVD):
disc
(Cd)
defragmenting HDDs
takes the fragmented files and rearranges the segments so that they run contiguously
A fragmented HDD is a slow HDD because the ‘drive heads’ (like the arm on an old record player) has to move to different parts of the disk to read a single file that has been broken into chunks. If a file is split into several places, then this will have an impact on the computer’s performance.
key features of magnetic storage
High storage capacity. 8 TiB HDD can be bought in 2023. Magnetic tapes are still sometimes used to back-up an organisation’s data.
Cheap per GB compared to the more expensive SSDs.
Reliable – the data does not degrade for a long time.
Has moving parts, so:
So not as portable as SSD (e.g. more likely to be affected by being dropped etc.).
Can consume more power than SSD.
Fragmentation reduces performance, unlike with a SSD.
how do optical discs work
An optical disc has several layers. There is always a thin aluminium reflective layeron the disc’s surface.
Each disc has a continuous track of data arranged in a spiral, rather than a series of concentric tracks, like on a hard disk.
Data is stored in the form ofpits (troughs)andlands (peaks)in the structure of the disc.
To read data, a laser is shone onto the surface of the disc. The pits and lands reflect light differently. The optical drive interprets the changes in reflected light as 1s and 0s.
are optical discs read-only
Some discs are read-only, but, if the disc allows writing, the laser burns new troughs into the reflective surface of the disc. The laser is set to a higher intensity than it is while reading data.
RW discs use a chemical dye that allows the burning process to be reversed, allowing you to rewrite the data on a disc.
3 different forms of optical discs
ROM discs areread-onlyversions and were primarily used to distribute data such as movies, music, and software.
R stands forrecordable. These discs can be written to once and then only read from that point on. They were typically used to archive data.
RW meansrewritable. These discs can be written and read many times.
key features of optical storage
Low storage capacity compared to other types …
… but still cheap per gigabyte.
Slow to access data.
Slow read / write speed
Has moving parts, which eventually fail. Easily damaged when dropped.
Thin, lightweight, portable (but a little fragile – easily scratched).
Good for long-term storage (e.g. archiving) as the data on the disc does not degrade quickly.
when HDDs are read…
The arm moves across to be above the right track.
The required sector comes around under the head.
The magnetised surface induces a tiny current in the head.
The disk controller translates this into 1s and 0s.
Steps 1 and 2 involve physical movement, which takes time. Data, therefore, does not come from the disk immediately.
when optical discs are read…
The disc spins.
The tracking mechanism moves the laser into the correct position over the disc.
The laser shines on to the disc and is reflected back on to a light sensor. When the laser shines on the disc surface, lands reflect the light back, whereas pits scatter the laser beam.
Signals from the sensor are translated into 1s and 0s. Reflected light (lands) represents a binary ‘1’, and no reflection (pits) represents a binary ‘0’.
1 = best, 3 = worst
