Memory and storage
Memory and storage devices can be split up into three distinct groups:
• primary memory
• secondary storage
• off-line storage.
Below diagram shows how these all link together:
Primary memory
Random Access Memory (RAM)
The features of RANDOM ACCESS MEMORY (RAM) are:
★ It is volatile/temporary memory (the contents of the memory are lost when the power to the RAM is turned off)
★ It is used to store:
• data,
• files,
• part of the operating system that are currently in use it can be written to or read from and the contents of the memory can be changed.
In general, the larger the size of RAM the faster the computer will operate.
In reality, the RAM never runs out of memory; it continues to operate but just gets slower and slower. As the RAM becomes full, the processor has to continually access the hard disk drive to overwrite old data on RAM with new data. By increasing the RAM size, the number of times this access operation is carried out is reduced, making the computer run faster.
RAM is much faster to write to or read from than other types of memory; but its main drawback is its volatility.
RAM is further divided into Two major technology
• dynamic ram (DRAM )
• static RAM (SRAM )
a)Dynamic RAM (DRAM)
Each DYNAMIC RAM (DRAM) chip consists of a number of transistors and
capacitors. Each of these parts is tiny since a single RAM chip will contain
millions of transistors and capacitors. The function of each part is:
• capacitor - this holds the bits of information (0 or 1)
• transistor - this acts like a S\llitch; it allows tl1e chip control circuitry to read the
capacitor or change the capacitor's value.
This type of RAM needs to be constantly REFRESHED ( that is, the capacitor needs
to be recharged every 15 microseconds otherwise it would lose its value).
If it wasn't refreshed, the capacitor's charge would leak away very quickly, leaving
every capacitor with the value 0.
DRAMs have a number of advantages over SRAMs:
• they are much less expensive to manufacture than SRAM
• they consume less power than SRAM
• they have a higher storage capacity than SRAM.
b)Static RAM (SRAM)
A big difference between SRAM and DRAM is that this type of memory doesn't
need to be constantly refreshed.
It makes use of 'flip flops' which hold each bit of memory.
SRAM is much faster than DRAM when it comes to data access (typically,
access time for SRAM is 25 nanoseconds and for DRAM is 60 nanoseconds).
DRAM is the most common type of RAM used in computers, but where
absolute speed is essential, then SRAM is the preferred technology. For example,
the processor's MEMORY CACHE is the high speed portion of the memory; it is
effective because most programs access the same data or instructions many times.
By keeping as much of this information as possible in SRAM, the computer avoids
having to access the slower DRAM.
Read Only Memory (ROM)
The main features of READ ONLY MEMORY (ROM) can be summarised as follows:
• they are non-volatile/permanent memories (the contents of the memory
remain even when the power to the ROM is turned off)
• they are often used to store the start-up instructions when the computer is first
switched on (for example, ROM might store the basic input/output system
(BIOS ))
• the data or contents of a ROM chip can only be read; they cannot be changed.
SECONDARY STORAGE
Hard Disk Drive (HDD)
Hard Disk Drive (HDD) are probably still the most common method to used to store data in a computer.
Data is stored in a digital format on the magnetic surfaces of the disks (or platters, as they are frequently called). The hard disk drive will have a number of platters which can spin at about 7000 times a second. A number of read-write heads can access all of the surfaces in the disk drive. Normally each platter will have two surfaces which can be used to store the data. These read-write heads can move very quickly - typically they can move from the centre of the disk to the edge of the disk (and back again) 50 times a second. Data is stored on the surface in sectors and tracks
A sector on a given track will contain a fixed number of bytes. Unfortunately, hard disk drives have very slow data access when compared to, for example, RAM. Many applications require the read-write heads to constantly seek for the correct blocks of data; this means a large number of head movements. The effects of LATENCY then become very significant. Latency is defined as the time it takes for a specific block of data on a data track to rotate around to the read-write head.
Users will sometimes notice the effect of latency when they see messages such as 'please wait' or, at its worst, 'not responding".
6 MEMORY AND DATA STORAGE
Latency is an issue in HDDs as described earlier. SOLID-STATE DRIVES (SSD) remove this issue considerably. They have no moving parts and all data is retrieved at the same rate. They don't rely on magnetic properties; the most common type of solid-state storage devices store data by controlling the movement of electrons within NAND chips. The data is stored as Os and 1s in millions of tiny transistors within the chip. This effectively produces a non-volatile rewritable memory.
However, a number of solid-state storage devices sometimes use ELECTRONICALLY ERASABLE PROGRAMMABLE READ-ONLY MEMORY (EEPROM) technology. The main difference is the use of NOR chips rather than NAND. This makes them faster in operation but devices using EEPROM are considerably more expensive than those that use NAND technology.
EEPROM also allows data to be read or erased in single bytes at a time.
Use ofNAND only allows blocks of data to be read or erased.
This makes EEPROM technology more useful in certain applications where data needs to be accessed or erased in byte-sized chunks. Because of the cost implications, the majority of solid-state storage devices use NAND technology. The two types are usually distinguished by the terms FLASH (use NAND) and EEPROM (use NOR).
So what are the main benefits of using SSD rather than HDD?
The mainbenefits of SSDs are summarised below:
• they are more reliable (no moving parts to go wrong)
• they are considerably lighter (which makes them suitable for laptops)
• they don't have to 'get up to speed' before they work properly • they have a lower power consumption
• they run much cooler than HDDs (these last two points again make suitable for laptop computers)
• because they have no moving parts, they are very thin
• data access is considerably faster than HDD.
The main drawback of SSD is the questionable longevity of the technology. Most solid state storage devices are conservatively rated at only 20 GB write operations per day over a three-year period - this is known as SSD endurance. For this reason, SSD technology is not used in servers, for example, where a huge number of write operations take place every day. However, this issue is being addressed by a number of manufacturers to improve the durability of these solid-state systems.
test2
ReplyDelete