On a computer that has no stored programs before--eniac
ENIAC 30.48 meters long, 1 meters wide, 2.4 meters high, covers an area of about 170 square meters, 30 operating stations, weighing up to 30 long ton, power consumption 150-kilowatt, the cost of 480,000 U.S. dollars. It contains 17,468 vacuum tubes 7,200 crystal diodes, 1,500 medium turn, 70,000 resistors, 10,000 capacitors, 1500 relays, 6,000 multiple switches, 5,000 additions per second or 400 multiplication, is 1000 times times the relay computer, 200,000 times times the manual calculation.
There are two problems with the original ENIAC,
there is no memory and it is controlled by a wiring board .its storage capacity is very small, can only store 20 words of 10-bit decimal number, so can only use the method of wiring connection programming, each problem must
rely on artificial to change the line ,even after a couple of days, the calculation speed is offset by this work.
The beginning of a new era - the proposed storage program most of the current computers are
von Neumann computers, which conform to the storage program concept of the computer, have the following 3 features:
(1) The computer hardware should be composed of the arithmetic, memory, controller, input device and output device 5 basic type parts. (2) The internal computer uses the binary to represent the instruction and the data. (3) put the programmed and raw data into the memory first, and then start the computer work, this is the basic meaning of the storage program.
main parts of the computerthe 5 basic parts of a computer--arithmetic, memory, controller, input device, output device.
(1)input Devicethe task of the input device is to send the program and data compiled by people to the computer, and convert them into information that can be recognized and accepted within the computer.
(2) output devicethe task of the output device is to output the computer's processing results in the form of numbers, text graphics, images, sounds, and so on.
(3) Memorymemory is a component used to store programs and data, and is the basis for computer-controlled storage.
Tiered Storage Systems:
Main Memory: We generally call the memory, is the main memory, can be directly accessed by the CPU, storage speed, but small capacity, generally used to store the program currently executing programs and data. Auxiliary Memory: Now the hard disk and so on are auxiliary memory, this kind of memory is set on the outside of the host (the host refers to the CPU plus the main memory), storage capacity, low price, but the slowest storage, generally used to store temporarily do not participate in running programs and data. The CPU does not have direct access to the secondary storage, and the program and data are transmitted to main memory when needed.
Cache Memory: CPU speed is still much higher than the storage speed of main memory (now higher DDR3 1600MHz memory and high CPU clock is 3.3GHz), so in main memory and the CPU to add a high-speed buffer memory, cache access faster than main memory, but smaller than memory, the price is more expensive, to store the most urgently needed processing of the program and data, in order to quickly provide instructions and data to the CPU.
(4) Computing devicethe task of an operator is to process and manipulate information, in short, to
perform arithmetic and logical operations , so it is called an arithmetic logic operation part, an ALU. The
core is the adder , and there are a number of
general-purpose registers or accumulator registers in the op-amp. Note: Registers have a much faster access speed than memory.
(5) ControllerThe main work of the controller is
to take instructions, translate the instructions, control the computer parts of the implementation of the tasks prescribed by the instructions.
For
more information about caching, refer to the content: L1 cache (cache) is the first CPU cache, which is divided into data cache and instruction cache. The capacity and structure of the built-in L1 cache has a large impact on the performance of the CPU, but the buffer memory is composed of static RAM, the structure is more complex, the capacity of the L1 cache cannot be too large in the case that the CPU core area is not too large. The L1 cache capacity of the general server CPU is usually 32-4096kb. L2 due to the limitation of L1 cache capacity, in order to increase the CPU speed again, place a high-speed memory outside the CPU, that is, level two cache. The operating frequency is more flexible, can be the same frequency with the CPU, can also be different. When the CPU reads the data, it looks for the L1, then the L2, then the memory, and then the external memory. Therefore, the impact of L2 on the system can not be ignored. L3 is now built-in. What it does, however, is that the application of the L3 cache can further reduce memory latency while improving the performance of the processor when large data volumes are computed. Reduce memory latency and increase the ability to compute large data volumes are useful for games. The increased L3 cache in the server domain still has a significant performance boost. For example, a configuration with a larger L3 cache can be more efficient with physical memory, so it is slower for the disk I/O subsystem to handle more data requests. Processors with larger L3 caches provide more efficient file system cache behavior and shorter message and processor queue lengths. that's all for today.
This article link: http://www.cnblogs.com/cposture/p/4311349.html
"Original" Storage program and von Neumann type computer
