Cache

At present, almost all the equipment that we usually use has a digital similar to that of a PC , whether in mobile phones , video game consoles or even new generation smart televisions. These computers have processors powered by processing cores, RAM cards, and other components that help the computers complete tasks with regard to digital processes .

Definition

The cache memory within computing is an electronic component that is found both in the hardware and in the software , it is responsible for storing recurring data so that it is easily accessible and faster to the requests generated by the system.

The cache is taken as a buffer special memory that computers have all this plays similar functions as the main memory. One of the most recognized caches are those of Internet browsers which maintain temporary downloads made from the Internet to have the information available for the internal system.

Characteristics

• It allows you to quickly and organized certain data, readings or files in common use, or that have recently been added as downloads from the internet.
• Presents an order system according to the importance of the file according to the requirement of each operating system
• It allows the processor to improve its performance and the results of the tasks, arranging the information from it as if it were a tool for continuous use.
• It relieves the internal processes of random storage memories , better known as RAM .
• Despite being an important part of the operating system, it does not take up a lot of space within the hardware or software.

What is cache memory for?

Within the cache memory system there are two main functions:

• The first is to keep files in an organized and relevant way, temporarily but relevant to the system when completing tasks required of the computer .
• The second task of the cache memory is to unload part of the information stored in the main memory, so that there are no failures when exchanging information due to content oversaturation.

History

The origin of cache memory was born when the first computers, as far as memory is concerned, could no longer cover the need for processors which were working at a higher speed than that of memories.

Therefore, the system engineers decide to incorporate a small auxiliary memory which would serve to assist the microprocessor, thus reducing the waiting time for data recovery.

The term was born in England where it is given the name of ” Cache ” which translates as a hidden place or hiding place to keep information or contraband.

How a cache works

The first thing that the cache memory is in charge of doing is organizing the information at different levels, these go from the lowest to the highest depending on how far they are from the processor . It is organized in 3 levels to facilitate the work of the processors when collecting the information.

When the processor needs information, it first resorts to the cache, there it is in charge of searching for each level until it finds the one required. In case it is not there, the processor looks for it in the internal memory and the cache saves data from the search to evaluate if it is necessary to incorporate it into the cache.

All levels within the cache have control centers where all the information handled is prepared and as the level increases ( from 3rd to 1st ) the amount of memory information is greater .

Cache memory types

Within the cache memories there are 3 that are the most used:

• Dedicated memory: the one to which a part of the RAM is composed , this usually stores data that was recently read in order to speed up the loads within them.
• Track cache: it is a solid composition memory of the RAM type, this is found mostly in supercomputers due to the high costs of these parts.
• Web cache: this is responsible for saving web- type documents in order to reduce time consuming processes such as downloads, server oversaturation and broadband expenses.

Velocity

The speed inside the cache memories is measured in “ns” which translates into nano seconds, the cache memory handles response times of between 15 and 35 ns to the processor, but this increases or decreases according to the number of gigabytes and also the brands of RAM cards, which is where the cache is mainly located in computers.

Advantage

• Unloads all data storage processes within random access memory.
• Increases the response speed of the processor when executing processes and providing responses.
• In the area of ​​internet it allows you to access the files just viewed or uploaded without consuming data or extra charges to the rent of the broadband.

Disadvantages

When it is not properly optimized , the cache memory causes problems in the processors, hindering the tasks that must be carried out.

The cache memory in some cases fails to erase its temporary files, becoming overloaded with information and slowing down all the information exchange processes between the cache and the main memory, completely affecting the computer.

Cache memory optimization

The cache memories must be optimized in order to reduce the failures in it known as “miss rate” , also reduce the penalty for failures of the same and finally reduce the access time to the time of hits in the memory “hit time “ .

The most common failures due to card problems are:

• Forced failures where access to a block of information that is not in the cache occurs.
• Due to low capacity where memory does not have enough space to cover the necessary blocks in the execution of a program.
• By conflict between the blocks that are disorganized within the memory .

The most used techniques to reduce misses within the cache memory are the following:

• Increasing the block sizes in order to avoid forced failures , however, in the long run can present problems with conflict failures.
• Increasing the associative method is another widely used method, but it ends up increasing the response time within memory.
• Add a victim cache in order to store blocks discarded due to capacity failures or conflict.
• Improve the information compiler so that it improves the data ordering system so that there are no conflicts between the different blocks.