Processor Flashcards
CPU
central processing unit- the brains of the computer where most calculations take place. Data is processed and programs are executed- most important element of a computer system
what is CPU made up of
ALU, buses, dedicated registers, control unit
control unit
uses electrical signals to control and coordinate the activities and timing of the CPU, directing THE FLOW OF DATA between the CPU and other devices. It accepts the next instruction, decodes it into sequential steps and manages the executing including fetching data and addresses from registers, storing the resulting data in memory registers.
buses
a set of parallel wires of 8,16,32,64 signal lines connecting 2 or more components of a computer. They connect processor, memory and IO controllers, transferring data and signals/one part of the motherboard to another They are a shared transmission medium and only one device can transmit along a bus at any one time. control, data and address bus makes up system bus
control bus
a BIDIRECTIONAL bus sending control signals. it ensures the use and access of the other two buses by different SHARED components doesn’t mean conflict. Transmits command, timing and specific status information between system components
bus request signal
indicates a device is requesting the use of the data bus
bus grant signal
indicates the CPU has granted access to data bus
memory write
causes data in the data bus to be written into an addressed location
memory read
causes data from addressed location to be placed onto the data bus
*interrupt request
indicates a device is requesting access to the CPU
clock
used to synchronise operations
data bus
provides a bi directional path for moving data and instructions between system COMPONENTS
address bus
single direction: processor to memory/IO device. each word has its own specific address so memory address of words (used as operands in program instructions) are transmitted so data can be retrieved and sent back to the processor. Results of instruction transmitted via data bus to location
word*
memory is divided up internally into units called words. they’re a fixed size group of DIGITS: 16,32,64 bits (different processor type: different word size). Handled as a unit by the processor
ALU
performes arithmetic and logic operations on data: e.g. add, subtract, divide, multiply on fixed and floating point, shift operations (shuffling left and right), boolean: comparing, and, or, not, xor
registers*
special memory cells operating at very high speed. up to 16 general purpose in a CPU: where all ARITHMETIC, LOGIC, SHIFT operations occur
accumulator (register)*
some special purpose CPUs use a SINGLE ACCUM instead of general purpose registers to simplify design. Calculations take place here and the results of the latest calculation are held here
PC (register)
holds address of the next instruction to be executed (next instruction in sequence or address to jump to if current instruction is to jump to another. Is INCREMENTED when read. Determines SEQUENCE of instructions being read/executed
CIR
holds current instruction currently being executed, split into the opcode and operand
*MAR
hold address of the memory location from which data or instructions are to be fetched or to which data is to be written, carried by the ADDRESS BUS
MDR/MBR
data used to temporarily store data read/written to memory on the data bus from/to the processor
FDE
the sequence of operations involved in executing an instruction, the cycle is repeated as each instruction of the program is executed
fetch
retrieving an instruction from main memory
what happens during fetch
address of next instruction copied from PC to MAR
address sent down address bus
read instruction command sent down CONTROL bus
instruction sent down data bus and copied to MDR
PC incremented TO hold address of next instruction
contents of MDR copied to CIR
decode
determines what the instruction is
what happens in decoding
instruction held in CIR is split into OPCODE and OPERAND
OPCODE (basic operation and addressing mode) determines TYPE of instruction and HARDWARE needed to execute it and OPERAND holds address of data to be used with the operation which is copied to the MAR/ holds actual data to be operated on, copied to MDR and may be passed onto the ALU/accumulator
execute
carrying out instruction
what happens during execution
appropriate instruction/opcode is carried out on operand. result is stored in appropriate register. cpu sets up to begin another cycle
factors affecting processor performance
clock speed
number of cores/duplicate processors linked together on a single chip
amount and type of cache
system clock
generates a series of signals, switching between 1 and 0 millions of times per second, synchronising CPU operations
system clock and CPU performance
CPU continually runs FE cycle, speed is governed by system clock/clock chip.
each CPU starts as clock changes from 0 to 1 most of the time
operations can’t be performed faster than clock speed but some may take longer than one clock cycle
the faster the clock speed the faster the instructions executed
1Hz v 1 GHz
1 clock cycle per sec v 1 billion cycles per sec
typical PC speed
2-4 GHz
higher clock speed?
faster performance, hotter, more power consumed/overclocking, expensive
lower clock speed
lower performance
needs less power (good battery life)
less costly
LMC number of instructions, memory locations and max data value
and number of bits in one word
11 instructions, 100 memory locations, 999 greatest data value
16 bit word
secondary storage device
a peripheral/external device which holds data and programs for a long period of time, even when switched off
RAM adv
quick access, allow changes to be made, allow files to be used
what to take into account for storage devices
cost, speed, capacity, portability
Secondary storage disadv
can’t be accessed directly by processor, slower retrieval speeds
all storage devices
create and maintain a toggle state (1/0) without power
magnetic storage devices
magnetic material eg ferrous particles on a disk or tape, polarised to represent 1 or 0, storing BINARY data eg hard disk drive
hard disk drive uses
in a typical PC main storage, storing OS, application software and files
adv hard disk
huge capacity, 500GB+: smaller denser s.a means 3.5in can now hold 640gb, fast access over 1000Kb/sec, random access- can be read directly from any part
disadv hard disk
not portable as large and moving parts, data lost if fail and no backup
hard disk
high speed spinning rigid (10,000RPM) disks/platters (with own drive heads), coated with magnetic ferrous particles, polarised 1/0 storing binary data. split into concentric circles and divided into tracks and sectors. drive head can access different parts and passes over parts to read/write into. drive head parked to side when not in use to prevent damage from movement.
optical disk uses
most application programs and games come on CDs, DVDs, blu-ray
adv optical disk
cheap, hold reasonable amount (800MB CD, 4.7GB DVD, 50GB), portable and drives are common in PCs, fast data transfer rates (88.5Mbit/sec)
disadv optical disk
scratched, can’t alway overwrite, storage isn’t amazing
How optical disks work
Using a high powered laser to burn (change chem properties of) sections of its surface making them less reflective. A laser at a lower point reads the disk by shining light onto the surface and a sensor measures how much is reflected back. Pits and lands are non reflective and reflective = 0 and 1. One single track as a tight spiral.
ROM 700MB and BluRay 50GB because shorter wavelength laser =
smaller pits so more to fit in same space and track tightly wound so much longer.
Recordable = reflective layer with transparent dye coating, less reflective when spot laser burns a spot in the track
RW: laser and magnet to heat spot and set it as 0 or 1 with magnet before cools: phase change alloy changes between amorphous and crystalline states by changing power of beam
Solid state disks working
Storing binary data on semiconductor material. Different areas acting as switches. Array of chips arranged on a board in standard size housing to fit into PCs. Millions of NAND flash memory cells and controller to manage pages and blocks of memory. Currents on bit and word lines, activating movement of electrons from source to drain and word line current forces a few across through insulated oxide layer to floating gate, which are trapped once current off. Some charge trapped = 0, no charge =1. Memory stored in pages and grouped into blocks. Data erased as whole blocks
Adv SSD
Faster than hard disk: no need to move RW head, portable and light, consume less power so better battery life and cooler, more durable, silent, not affected by magnets
Disadv of SSD/flash
Data can’t be overwritten (has to be erased as blocks first)
Limited number of writes
Less capacity than hard disk drive
Why does RAM lead to better performance
More programs processed at the same time
Process larger files
No need to resort to VM
Input devices
Scanner Biometric devices Keyboard Mouse Touchscreen Sensors Barcode reader OMR OCR
Output devices
Lights, monitor, speakers, headphones, plotter, touchscreen
Define input device
Device allowing us to enter data into a computer system for processing
Define output device
Allowing us to EXTRACT data from a computer system, converting it to a form we can understand
concurrent processing
When several processes are running on a single processor, each given a time slice according to scheduling algorithms. It gives the appearance that several tasks are being performed simultaneously
adv and disadv of concurrent processing
increased throughput: number of tasks completed in a given time increases
time that would be wasted waited for the user to input data is used on another task
if large number of users trying to run programs and some involve a lot of computation, they will take longer to complete
adv and disadv of parallel programming
speeds up processing significantly esp when repetitive calculations need to be performed on large amounts of data
browsers can display several web pages in separate windows. different processors do different tasks in each window
graphics processors can quickly render a 3D object by working simultaneously on individual components
some tasks may run faster on a single processor, some cannot take advantage of more than one processor. there is overhead in coordinating the processors
