Basic trems used in CPU


1.  CU(Control Unit)
2. MU(Memory Unit)
3.ALU(Arthnetic and logic unit)

1.The Control Unit (CU) is digital circuitry contained within the processor that coordinates the sequence of data movements into, out of, and between a processor's many sub-units. The result of these routed data movements through various digital circuits (sub-units) within the processor produces the manipulated data expected by a software instruction (loaded earlier, likely from memory). In a way, the CU is the "brain within the brain", as it controls (conducts) data flow inside the processor and additionally provides several external control signals to the rest of the computer to further direct data and instructions to/from processor external destinations (i.e. memory).
Examples of devices that require a CU are CPUs and graphics processing units (GPUs). The CU receives external instructions or commands which it converts into a sequence of control signals that the CU applies to the data path to implement a sequence of register-transfer level operations.[3]
More precisely, the Control Unit (CU) is generally a sizable collection of complex digital circuitry interconnecting and controlling the many execution units (i.e. ALU, data buffers, registers) contained within a CPU. The CU is normally the first CPU unit to accept from an externally stored computer program, a single instruction (based on the CPU's instruction set). The CU then decodes this individual instruction into several sequential steps (fetching addresses/data from registers/memory, managing execution [i.e. data sent to the ALU or I/O], and storing the resulting data back into registers/memory) that controls and coordinates the CPU's inner works to properly manipulate the data. The design of these sequential steps are based on the needs of each instruction and can range in number of steps, the order of execution, and which units are enabled. Thus by only using a program of set instructions in memory, the CU will configure all the CPU's data flows as needed to manipulate the data correctly between instructions. This results in a computer that could run a complete program and requiring no human intervention to make hardware changes between instructions (as had to be done when using only punch cards for computations before stored programmed computers with CUs where invented). These detailed steps from the CU dictate which of the CPU's interconnecting hardware control signals to enable/disable or which CPU units are selected/de-selected and the unit's proper order of execution as required by the instruction's operation to produce the desired manipulated data. Additionally, the CU's orderly hardware coordination properly sequences these control signals then configures the many hardware units comprising the CPU, directing how data should also be moved, changed, and stored outside the CPU (i.e. memory) according to the instruction's objective. Depending on the type of instruction entering the CU, the order and number of sequential steps produced by the CU could vary the selection and configuration of which parts of the CPU's hardware are utilized to achieve the instruction's objective (mainly moving, storing, and modifying data within the CPU). This one feature, that efficiently uses just software instructions to control/select/configure a computer's CPU hardware (via the CU) and eventually manipulates a program's data, is a significant reason most modern computers are flexible and universal when running various programs.[citation needed] As compared to some 1930s or 1940s computers without a proper CU, they often required rewiring their hardware when changing programs. This CU instruction decode process is then repeated when the Program Counter is incremented to the next stored program address and the new instruction enters the CU from that address, and so on till the programs end.
Other more advanced forms of Control Units manage the translation of instructions (but not the data containing portion) into several micro-instructions and the CU manages the scheduling of the micro-instructions between the selected execution units to which the data is then channeled and changed according to the execution unit's function (i.e., ALU contains several functions). On some processors, the Control Unit may be further broken down into additional units, such as an instruction unit or scheduling unit to handle scheduling, or a retirement unit to deal with results coming from the instruction pipeline.Again, the Control Unit orchestrates the main functions of the CPU: carrying out stored instructions in the software program then directing the flow of data throughout the computer based upon these instructions (roughly likened to how traffic lights will systematically control the flow of cars [containing data] to different locations within the traffic grid [CPU] until it parks at the desired parking spot [memory address/register]. The car occupants [data] then go into the building [execution unit] and comes back changed in some way then get back into the car and returns to another location via the controlled traffic grid).
2 ALU : ALU Stands for "Arithmetic Logic Unit." An ALU is an integrated CIRCUIT within a CPU or GPU that performs arithmetic and logic operations. Arithmetic instructions include addition, subtraction, and shifting operations, while logic instructions include booleean comparisons, such as AND, OR, XOR, and NOT operations.
ALUs are designed to perform integer calculations. Therefore, besides adding and subtracting numbers, ALUs often handle the multiplication of two integers, since the result is also an integer. However, ALUs typically do not perform division operations, since the result may be a fraction, or a "floating point" number. Instead, division operations are usually handled by the floating-point unit (FFU), which also performs other non-integer calculations.
While the ALU is a fundamental component of all processor, the design and function of an ALU may vary between different processor models. For example, some ALUs only perform integer calculations, while others are designed to handle floating point operations as well. Some processors contain a single ALU, while others include several arithmetic logic units that work together to perform calculations. Regardless of the way an ALU is designed, its primary job is to handle integer operations. Therefore, a computer's integer performance is tied directly to the processing speed of the ALU.

3.Memory Unit:(storage unit), a unit of a computer or an independent device designed to record, store, and reproduce information.
Memory units are used most widely in digital computers but also have applications in automation, remote-control, nuclear-physics, and other devices, where they store discrete (for the most part) information, coordinate in time the operation ofseveral facilities, or accumulate data to be transmitted through remote-control channels. The recording of information inmemory units is based on various physical principles: the mechanical displacement or removal of part of the material of theinformation carrier (paper tape or punchcards), a change in the magnetic state of the material (magnetic tapes, disks,drums, and ferrite cores), the accumulation of an electrostatic charge in dielectrics (condensor memory units andcathoderay memory tubes), audio or ultrasonic oscillations (delay lines), and the application of superconductivity (cryogeniccomponents).
The main indexes defining the efficiency of memory units are the capacity Mthe maximum number of words or symbolsthat can be simultaneously stored in a memory unit, expressed in binary units (bits) or bytes (eight bits); and the speed,defined by the duration of a full access cycle to the memory unit Tc (or sometimes by the readout time) or by the accessrate F = 1/TCSometimes a generalized parameter—the information capacity W = M x F (reaching 1013 bytes/sec in thebest modern memory units)—is used to characterize a memory unit.
According to their intended use, methods of data distribution, and special characteristics of operation, memory units aregenerally classified as shown in Figure 1. Depending on the type of retrieval of desired information, a distinction is madebetween addressed memory units, in which a specific number (address) is assigned to each memory cell and the requiredinformation is sought at a specific address, and associative memory units, in which the information is sought on the basis ofa set of indicators. Either sequential or cyclic access to cells or random access, in which a cell is accessed independentlyof its location among other cells, is possible in a memory unit.
Depending on the frequency of access, memory units are divided into nonerasable units, which allow single recording withsubsequent repeated readout without regeneration (such as diode matrices, punchcards, and paper tapes), and erasableunits (memory units on magnetic carriers, ferrite cores, flip-flops, and other devices).
Memory units in which the states of the carrier that correspond to the assigned code are fixed with respect to theinformation carrier are called static memories. All memory units with nondestructive readout are also classified as staticunits. In dynamic memory units a sequence of signals that correspond to the fixed code circulates through a closed loopthat includes a delay line. Static memory units may be non-volatile, in which case the information is retained for anindefinitely long period (such as flip-flop memory units or memory units with ferrite cores), or volatile, in which case theyhave the property of spontaneous erasure of information (condensor memory units and cathoderay memory tubes).
To store large data arrays so-called external memory units are most often used, the recording being accomplished onmagnetic carriers: magnetic tapes, drums, and disks. By means of parallel switching of several blocks of the memory unit itis theoretically possible to store unlimited volumes of information. The capacity of modern external memory units usingmagnetic tapes runs as high as 108 bytes (with up to 256 on-line blocks); the capacity of those using magnetic disks is upto 6 x 108 bytes. The input (output) rate is 3.2 x 105 bytes/sec for magnetic-tape memory units and 2 x 105 bytes/sec formagnetic disk memory units.
Working storage units, which are generally constructed with toroidal ferrite cores (up to 95 percent of all working storageunits) or less often with other ferromagnetic components (such as multiaperture ferrite plates and magnetic thin films), areused in digital computers to store data needed in the next stages of problem-solving. Integrated-circuit memory units areconsidered to be extremely promising. The capacity of working storage units in modern large digital computers runs as highas 16 x 106 bytes; the write and read (access) time runs from hundredths of a microsecond to several microseconds.

Previous
Next Post »