System Memory
System
Memory (Memory) are electronic components that store instructions
waiting to be executed by the processor, data needed by insruksi
(command) and the results of the processed data (information). Memory chip typically consists of one or more of the other circuits in the processor board. Computer
memory can be described as a whiteboard, where everyone enters the room
can read and utilize the existing data without changing the composition
presented. The
data are processed by a computer, is still stored in the memory, and in
this case the computer only reads the data and then process it. One time the data stored in computer memory, the data will remain there forever. Each time memory is full, then the data can be removed in part or in whole to be replaced with new data.Characteristics of memory systems in general:a. Location• CPUThis built-in memory is the CPU (microprocessor) and is required for all activities of the CPU. This memory called registers.• Internal (main)This
memory outside the processor chip but internal to the computer system
and is required by the CPU to process execution (operations) program, to
be accessed directly by the processor (CPU) without any intermediate
module. The internal memory is often also referred to as primary memory or main memory. The internal memory is typically used RAM media• External (secondary)This
memory is external to the computer system and certainly beyond the CPU
and is required to store data or instructions permanently. Memory, it is not necessary in the execution process so it can not be accessed directly by the processor (CPU). For external memory access by the CPU to go through the controller / module I / O. External memory is often also referred to as secondary memory. This memory consists of peripherals such as storage devices: disks, magnetic tape, etc..a. Capacity• Size of wordInternal and external memory capacity is usually expressed in terms of bytes (1 byte = 8 bits) or a word.• The number of wordWord length is generally 8, 16, 32 bits.
b. Transfer UnitUnit equal to the number of channels transfer data into and out of the memory module. The concept of transfer units are:• Word, a unit of the "natural" memory organization. Word size is usually equal to the number of bits used for the representation of the number and length of instruction.• Addressable units, the number of systems, units are adressable word. But there is a system by addressing the byte level. In all cases the relationship between the length of a an address and the number N adressable unit is 2A = N.• Unit of transfer, is the number of bits that are read or written into the memory at a time. In external memory, data transfer is usually larger than a word, called a block.
c. Access MethodsThere are four types of data access unit, as follows.:• Sequential accessMemory is organized into units of data, called records. Access is made in the form of specific linear sequence. Addressing information used to separate records and to assist in the search. Mechanism of read / write are shared (shared read / write mechanism), by walking towards the desired location to remove record. Record access time varies. Examples of sequential access is the access to magnetic tape.• Direct accessSuch as sequential access, direct access also make use of shared read / write mechanism, but each block and the record has a unique address based on physical location. Access is made directly to the public range (general vicinity) to reach the final location. Access time varies. Examples of direct access is the access to the disk.• Random accessEach location can be selected at random and accessed and addressed directly. The time to access a particular location does not depend on the sequence of previous access and is constant. An example is a system random access main memory.• Associative accessEach word can be searched based on its content and not by address. As in RAM, each location has its own mechanism pengalamatannya. When the search was not constantly depend on the location or previous access patterns. An example is the associative cache memory access.a. PerformanceThere are three parameters for the performance of the memory system, namely:• Access time (Access Time)For RAM, access time is the time required to perform a read or write operation. As for non-RAM, access time is the time required to perform read and write mechanism at specific locations• Cycle time (Time Cycle)The cycle time is the access time plus the transient time until the signal is lost from the channel signal or to generate return data when the data is read destructively.• Transfer rate (transfer rate)The transfer rate is the speed of transfer of data to a memory unit or transferred from the memory unit. For RAM, the transfer rate is equal to 1 / (cycle time). Meanwhile, the non-RAM, apply the following equation.:
TN = average time to read / write the number of N bits.TA = average access timeN = number of bitsR = The transfer speed, in bits per second (bps)a. Type of PhysicalThere are two types of physical memory, namely:• semiconductor memoryThis memory technology using LSI or VLSI (very large scale integration). This memory is widely used for the internal memory as RAM.• Memory magnetic surfaceThis memory is widely used for external memory that is to disk or magnetic tape.b. Physical CharacteristicsThere are two criteria that reflect the physical characteristics of memory, namely:• Volatile and non-volatileIn volatile memory, the information will be naturally broken or lost when electrical power is turned off. In addition, the non-volatile memory, so the information recorded will still be there without damage prior to the change. In this memory power is not required to retain the information. The memory is non-volatile magnetic surface. Semiconductor memory can be either volatile or non-volatile.• Erasable and Non-erasableErasable means that memory contents can be removed and replaced with other information. Semiconductor memory that is not erased and is non-volatile ROM.1. Memory HierarchyThree questions in the design of memory, that is: How much? This involves kaspasitas. How soon? This is about the time of access, and how expensive are concerned the price? Each spectrum technology has the following relationship:• The smaller the access time, greater cost per bit.• The greater the capacity, the smaller the price per bit.• The greater the capacity, the greater the access time.To get the best performance, the memory must be able to follow the CPU. This means that when the CPU is executing instructions, we do not need to stop the CPU to wait for instructions or operands. As for getting the best performance, memory becomes expensive, berkasitas relatively low, and fast access time. To obtain optimal performance, it should be a combination of a memory component technology. From the combination of the memory hierarchy can be structured as follows:Goes down the hierarchy, then the things below will occur:a) Decrease in price per bitb) Capacity buildingc) Increasing access timed) Decrease the frequency of memory access by the CPU.The key to the success of this hierarchy is a reduction in the frequency of access. The slower memory then the CPU needs to access the less. Overall the computer system will be fixed quickly but needs a large memory capacity is met.
System
Memory (Memory) are electronic components that store instructions
waiting to be executed by the processor, data needed by insruksi
(command) and the results of the processed data (information). Memory chip typically consists of one or more of the other circuits in the processor board. Computer
memory can be described as a whiteboard, where everyone enters the room
can read and utilize the existing data without changing the composition
presented. The
data are processed by a computer, is still stored in the memory, and in
this case the computer only reads the data and then process it. One time the data stored in computer memory, the data will remain there forever. Each time memory is full, then the data can be removed in part or in whole to be replaced with new data.Characteristics of memory systems in general:a. Location• CPUThis built-in memory is the CPU (microprocessor) and is required for all activities of the CPU. This memory called registers.• Internal (main)This
memory outside the processor chip but internal to the computer system
and is required by the CPU to process execution (operations) program, to
be accessed directly by the processor (CPU) without any intermediate
module. The internal memory is often also referred to as primary memory or main memory. The internal memory is typically used RAM media• External (secondary)This
memory is external to the computer system and certainly beyond the CPU
and is required to store data or instructions permanently. Memory, it is not necessary in the execution process so it can not be accessed directly by the processor (CPU). For external memory access by the CPU to go through the controller / module I / O. External memory is often also referred to as secondary memory. This memory consists of peripherals such as storage devices: disks, magnetic tape, etc..a. Capacity• Size of wordInternal and external memory capacity is usually expressed in terms of bytes (1 byte = 8 bits) or a word.• The number of wordWord length is generally 8, 16, 32 bits.b. Transfer UnitUnit equal to the number of channels transfer data into and out of the memory module. The concept of transfer units are:• Word, a unit of the "natural" memory organization. Word size is usually equal to the number of bits used for the representation of the number and length of instruction.• Addressable units, the number of systems, units are adressable word. But there is a system by addressing the byte level. In all cases the relationship between the length of a an address and the number N adressable unit is 2A = N.• Unit of transfer, is the number of bits that are read or written into the memory at a time. In external memory, data transfer is usually larger than a word, called a block.
c. Access MethodsThere are four types of data access unit, as follows.:• Sequential accessMemory is organized into units of data, called records. Access is made in the form of specific linear sequence. Addressing information used to separate records and to assist in the search. Mechanism of read / write are shared (shared read / write mechanism), by walking towards the desired location to remove record. Record access time varies. Examples of sequential access is the access to magnetic tape.• Direct accessSuch as sequential access, direct access also make use of shared read / write mechanism, but each block and the record has a unique address based on physical location. Access is made directly to the public range (general vicinity) to reach the final location. Access time varies. Examples of direct access is the access to the disk.• Random accessEach location can be selected at random and accessed and addressed directly. The time to access a particular location does not depend on the sequence of previous access and is constant. An example is a system random access main memory.• Associative accessEach word can be searched based on its content and not by address. As in RAM, each location has its own mechanism pengalamatannya. When the search was not constantly depend on the location or previous access patterns. An example is the associative cache memory access.a. PerformanceThere are three parameters for the performance of the memory system, namely:• Access time (Access Time)For RAM, access time is the time required to perform a read or write operation. As for non-RAM, access time is the time required to perform read and write mechanism at specific locations• Cycle time (Time Cycle)The cycle time is the access time plus the transient time until the signal is lost from the channel signal or to generate return data when the data is read destructively.• Transfer rate (transfer rate)The transfer rate is the speed of transfer of data to a memory unit or transferred from the memory unit. For RAM, the transfer rate is equal to 1 / (cycle time). Meanwhile, the non-RAM, apply the following equation.:
TN = average time to read / write the number of N bits.TA = average access timeN = number of bitsR = The transfer speed, in bits per second (bps)a. Type of PhysicalThere are two types of physical memory, namely:• semiconductor memoryThis memory technology using LSI or VLSI (very large scale integration). This memory is widely used for the internal memory as RAM.• Memory magnetic surfaceThis memory is widely used for external memory that is to disk or magnetic tape.b. Physical CharacteristicsThere are two criteria that reflect the physical characteristics of memory, namely:• Volatile and non-volatileIn volatile memory, the information will be naturally broken or lost when electrical power is turned off. In addition, the non-volatile memory, so the information recorded will still be there without damage prior to the change. In this memory power is not required to retain the information. The memory is non-volatile magnetic surface. Semiconductor memory can be either volatile or non-volatile.• Erasable and Non-erasableErasable means that memory contents can be removed and replaced with other information. Semiconductor memory that is not erased and is non-volatile ROM.1. Memory HierarchyThree questions in the design of memory, that is: How much? This involves kaspasitas. How soon? This is about the time of access, and how expensive are concerned the price? Each spectrum technology has the following relationship:• The smaller the access time, greater cost per bit.• The greater the capacity, the smaller the price per bit.• The greater the capacity, the greater the access time.To get the best performance, the memory must be able to follow the CPU. This means that when the CPU is executing instructions, we do not need to stop the CPU to wait for instructions or operands. As for getting the best performance, memory becomes expensive, berkasitas relatively low, and fast access time. To obtain optimal performance, it should be a combination of a memory component technology. From the combination of the memory hierarchy can be structured as follows:Goes down the hierarchy, then the things below will occur:a) Decrease in price per bitb) Capacity buildingc) Increasing access timed) Decrease the frequency of memory access by the CPU.The key to the success of this hierarchy is a reduction in the frequency of access. The slower memory then the CPU needs to access the less. Overall the computer system will be fixed quickly but needs a large memory capacity is met.
Thanks for nice post.C programming details here
BalasHapus