Cache memory

Cache memory is static memory gets its name because it does not require the regeneration of both the operational dynamic memory. But regeneration can take up to 10% or more of CPU time. Save time when using a large cache, which allows it to operate at the processor frequency. Elementary cache consists of six transistors and is triggered reporting - a device that indefinitely retains one of its stable equilibrium states and abruptly switches on a signal from the outside from one state to another.

Cache memory is static memory gets its name because it does not require the regeneration of both the operational dynamic memory

For the first time the word "cache" was used in 1967 during the preparation of articles for publication in the journal "IBM Systems Journal". The article dealt with the memory improvements in the model 85 of the IBM System/360 series. The editor Lyle Johnson asked to come up with a more descriptive term than "fast buffer", but due to lack of ideas himself suggested the word "cache". The article was published in early 1968, the authors were awarded the IBM, their work has spread, and subsequently has been improved, and the word "cash" soon came to be used in the computer literature as a common term.

Why a cache? Since it operates at the processor speed, then the data loaded into it from the dynamic RAM memory. But not all, and those to which the processor often handles. Appeal to the processor cache, rather than the operative dynamic memory, greatly improves system performance.

However, the cache memory is limited in size. And so the processor does not always find the right data in it.

It should be noted that not all data can be read from the dynamic RAM in memory cache. There is a limit on the amount of cached memory and it depends on the used motherboard chipset card (in legacy systems) or on the type of processor (in modern systems).

There are two types of cache: a first layer and second layer. The first time cache of the first level came in 1986. It was a high-speed buffer between the CPU and RAM dynamic memory. Then we developed a cache of the second level, which was located on the motherboard and worked on the system bus frequency. Subsequently, the second level cache was put to a special fee (which is attached to the processor itself) and its data between the processor and transmitted on a separate processor bus. Now cache both species is within the processor.

Generally, the larger the cache, the higher the system performance. But by itself the volume does not guarantee high performance. For its, the main thing is the controller.