It is the principle of locality that gives us a chance to overcome the long late

Question

It is the principle of locality that gives us a chance to overcome the long latency of memory access. Please explain what are temporal locality and special locality, and give an very example fin the area of computer architecture) to illustrate the point. There are situations in pipelining when the next instruction cannot execute in the following clock cycle. These events are called hazards. And there are three different types. Please s what are they and give a slim example for each hazard to explain why. Could you briefly explain demonstrate Memory Hierarchy in a typical Pc setting.

Explanation / Answer

In a normal traditional computer we have many memories like

1.Secondary memory(hard disk )

2.primary memory(ram / rom)

3. on chip memory (cache/ registers)

Secondary memory:- secondary memory is generally a device which stores your data permanently. these types of memories are generally called as non- volatile memory. this type of memory are generally build on using flipflops by using the concepts of digital design.

Internal organisation of Secondary memory:

in internal memory of hard disk have a magnetic disk where the data is wriiten over it using a plotter and data is saved in the form of tracks/sectors .

these secondary memory devices are generally have good capacity of storage measuring from 500 mb to 1 tb but these hard disk are slow in acess speed the only reason why they are slow because they are not connected to the mortherboard directly and they are connected to the board using ata and sata cable which in turn reduces the speed of data transfer. apart from these the data transfer happens from ram to hard disk as the ram is primary memory already it will be busy with its tasks given by cpu so basing on the response time of the ram the speed gets reduced , to make sure the response time will not drag for more amount of time thier is method called DMA(direct memory acess) which will transfer data to hard disk instead of ram .but it is less expensive compared to other memory devices.

main memory : generally its is called as RAM (random acesss control) it is also known as volatile memory. in the whole computer it is the only component which directly communicates with the cpu. it is just like a heart of the computer . for any input operation or out put operation it needs to happen by ram.

input device---------->ram---------------->output device

suppose if you type in a letter on your keyboard it will go to ram from ram it will carry the data to the cpu, cpu will give you the instruction back to the ram then it will go back to the output device. so it acts a mediater to the cpu to the other devices to the computer. so cpu generally organise the tasks in the computer internally. when you are doing a work in ms word data will not get saved unless you hit on save button when you hit have button it willl move to hard disk or the data will save its data in ram for temporary basis. each ram has sizes like 500 mb, 1gb ,2gb etc it this space gets fillled up it will implement a concept called virtual memory(sharing an amount of hard disk memmory) and will save its data to make sure no data loss happens. rams are divided in to two types

1.static ram 2. dynamic ram

1.static ram is build of transistors . as rams are volatile memory it will forget data immediately once power is turnrd off so in staic memory it will forget or loose data immediately but in dynamic ram it is build by capacitors the data will sustain a period of time even the power is turned off. so this is the reason why dynamic rams are used extensively in computers now a days. its bit expensive than hard disk

cache memory is generally memory present in the chip or the processor itself so this memory is used to store the data which is used recently , it simply creates a index to make sure that regular data is refrenced directly without performing long search so that the speed of computation will be fast. generally the size of these cache memory are restricted to killobytes . they are three types of caches L1,L2,L3 to store data . its is costly than ram , hard disk.

apart from this computer has a another another on chip memory called register which is present inside the cpu it self it is generally used to perform internal computation within its memory instead of retriving data from other memory. its simply equivalent to the condition where you want to do 5+3 we can simply do it by our fingers but if it is 1500+3586 we used a pen to do it. so registers is like our fingers which is used to do computation within cpu itself with out referencing a memory. its size is generally in killobytes. its costly than ram, hardisk and cache

heiarchy of size and cost

1.hard disk ( giga bytes to tera bytes)

2. ram (megabytes to giga bytes)

3.cache ( kilobytes)

4. registers (bytes to killo bytes)

hiearchy of cost

1. register (very expensive)

2. cache ( expensive)

3.ram( a bit expensive)

4.hard disk( less expensive)

Related Questions

Navigate

• Browse (All)
• Browse I
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.