03-26-2011, 01:41 PM
(This post was last modified: 03-26-2011, 05:04 PM by Splodge0007.)
*Note, this tutorial may not be perfect as It's another way for me to recap/revise and see how much I've remembered/understood my computer architecture classes. If there is anything you think is wrong, please mention so, as I've written this whole tutorial myself.
Thank you for reading. This has actually helped me with some end of unit revision!
The processor
The processor, otherwise known as the brain of a computer is the central processing unit. It has many different "areas" in which an explanation into each part will be written below. Everything in your computer is relayed and controlled through the processor.
A brief explanation on what a processor functions as, is a complex piece of hardware in which reads and executes an instruction (In binary), then relaying them through various busses on the motherboard to the input-output system.
How it works
A request is given to the processor. The processor then pulls the instruction from the RAM. The processor then reads the instruction before sending it back to the RAM. The CPU then sends the data down the "System bus" to the appropriate sub-systems. After it has been sent to the sub-systems it is then sent back to the input-output system, and displayed on the monitor.
There are two types of ways a processor reads an instruction. These two methods are known as "Polling" and "Interrupt" requests.
What are Polling and Interrupt requests?
Like a human brain, a processor has a certain list of jobs to be done. This list is known as a polling sequence. The polling sequence is a sequence in which each task that the processor has to do is lined up in a logical order. Polling is for a single core in a processor, as each core has a separate polling sequence. Single core processors are slow due to taking so much time to do a task. Which is where multi-cores come into play. One core of the processor does a part of the task, then passes it onto another core and moves onto the next task. In a single core processor, it will do a part of the task, move onto another task and come back to do the original task. It does this so quickly that you cannot see that It's doing it, however It's the reason why single core processors are so slow.
An "Interrupt request" is where there is a logical polling sequence in place. When a "High-priority" task is required, an interrupt request is sent to the processor. What this does, is bypasses the polling sequence, jumping the task to the front of the queue. Thus meaning, It's quickly completed.
What are the individual parts in the processor?
A processor has multiple parts. These parts are :
The IR
The IR is known as the "Instruction Register". The IR basically receives an instruction and translates this instruction into binary so that the processor can understand the instruction.
The ALU
The ALU is known as the "Arithmetic Logic Unit". The ALU performs many mathematical and logical operations within a processor. It's also the final part of processing performed by the processor. When data has been sent out of the ALU it is loaded directly into computer memory. In some proccessors the ALU is actually split into two parts. Known as an AU and an LU (Arithmetic unit, Logic unit)
Cores
A modern day computer has more then one core. A common miss-conception is that multi-cores perform separate individual tasks. There are two ways in which cores operate.
The common way : As explained above, multiple cores help with the polling sequence in processing. As I've explained it above in quite some detail, I'll leave this part out.
The shared way : In multi-client system, the cores are split up and allocated to different users. This meaning, sharing resources. This means that each core is allocated to a particular user. This is more efficient in a large-scale server. This is because of interrupt requests being sent by many people, thus making the work-load heavier.
By splitting up the cores and allocating a space to each user, there are less overall interrupt requests. So that people experience a faster system. This method of processing is only usually seen in multi-client based operating systems such as Windows Server 2003/2008.
The cache
The cache is a small part of the processor. In simplistic terms, the cache is RAM inside the processor. Incredibly fast RAM at that. The cache is used to reduce the average time it takes the processor to access computer main memory. When a processor has to read or write to a location found inside the main memory. It first checks if there is a copy of such data inside It's own cache. If It's there, then the processor immediately reads or writes from It's own cache. This method is much faster then writing to the computers main memory.
What language does a processor understand?
A processor can only understand one language. This language is known as binary (As this tutorial isn't about binary, I won't go in-depth about it). Binary is a sequence of ones and zero's in which tells a computer "On or off".
Conclusion
So basically, from this tutorial you've possibly understood how a processor works. It receives an instruction, requests the instruction from the main memory, and then finally executes the instruction sending it back out to the main memory, and finally to the input/output system.
The processor, otherwise known as the brain of a computer is the central processing unit. It has many different "areas" in which an explanation into each part will be written below. Everything in your computer is relayed and controlled through the processor.
A brief explanation on what a processor functions as, is a complex piece of hardware in which reads and executes an instruction (In binary), then relaying them through various busses on the motherboard to the input-output system.
How it works
A request is given to the processor. The processor then pulls the instruction from the RAM. The processor then reads the instruction before sending it back to the RAM. The CPU then sends the data down the "System bus" to the appropriate sub-systems. After it has been sent to the sub-systems it is then sent back to the input-output system, and displayed on the monitor.
There are two types of ways a processor reads an instruction. These two methods are known as "Polling" and "Interrupt" requests.
What are Polling and Interrupt requests?
Like a human brain, a processor has a certain list of jobs to be done. This list is known as a polling sequence. The polling sequence is a sequence in which each task that the processor has to do is lined up in a logical order. Polling is for a single core in a processor, as each core has a separate polling sequence. Single core processors are slow due to taking so much time to do a task. Which is where multi-cores come into play. One core of the processor does a part of the task, then passes it onto another core and moves onto the next task. In a single core processor, it will do a part of the task, move onto another task and come back to do the original task. It does this so quickly that you cannot see that It's doing it, however It's the reason why single core processors are so slow.
An "Interrupt request" is where there is a logical polling sequence in place. When a "High-priority" task is required, an interrupt request is sent to the processor. What this does, is bypasses the polling sequence, jumping the task to the front of the queue. Thus meaning, It's quickly completed.
What are the individual parts in the processor?
A processor has multiple parts. These parts are :
- IR
- ALU
- Cores
- Cache
The IR
The IR is known as the "Instruction Register". The IR basically receives an instruction and translates this instruction into binary so that the processor can understand the instruction.
The ALU
The ALU is known as the "Arithmetic Logic Unit". The ALU performs many mathematical and logical operations within a processor. It's also the final part of processing performed by the processor. When data has been sent out of the ALU it is loaded directly into computer memory. In some proccessors the ALU is actually split into two parts. Known as an AU and an LU (Arithmetic unit, Logic unit)
Cores
A modern day computer has more then one core. A common miss-conception is that multi-cores perform separate individual tasks. There are two ways in which cores operate.
The common way : As explained above, multiple cores help with the polling sequence in processing. As I've explained it above in quite some detail, I'll leave this part out.
The shared way : In multi-client system, the cores are split up and allocated to different users. This meaning, sharing resources. This means that each core is allocated to a particular user. This is more efficient in a large-scale server. This is because of interrupt requests being sent by many people, thus making the work-load heavier.
By splitting up the cores and allocating a space to each user, there are less overall interrupt requests. So that people experience a faster system. This method of processing is only usually seen in multi-client based operating systems such as Windows Server 2003/2008.
The cache
The cache is a small part of the processor. In simplistic terms, the cache is RAM inside the processor. Incredibly fast RAM at that. The cache is used to reduce the average time it takes the processor to access computer main memory. When a processor has to read or write to a location found inside the main memory. It first checks if there is a copy of such data inside It's own cache. If It's there, then the processor immediately reads or writes from It's own cache. This method is much faster then writing to the computers main memory.
What language does a processor understand?
A processor can only understand one language. This language is known as binary (As this tutorial isn't about binary, I won't go in-depth about it). Binary is a sequence of ones and zero's in which tells a computer "On or off".
Conclusion
So basically, from this tutorial you've possibly understood how a processor works. It receives an instruction, requests the instruction from the main memory, and then finally executes the instruction sending it back out to the main memory, and finally to the input/output system.
Thank you for reading. This has actually helped me with some end of unit revision!
![[Image: IYOy9.png]](http://gfxf.net/images/IYOy9.png)
Maybe you could add in an explanation for the difference between dual/quad core processors as well? Just something I didn't see explained in the post.
