Digital-electronics 简明教程

Cache Memory Design

Cache memory 是一种可以加速 CPU 的高速半导体存储器。它充当了 CPU 和主存储器之间的缓冲器。它用于保存 CPU 最常使用的那些部分数据和程序。在操作系统将数据和程序的部分从磁盘传输到高速缓存存储器时，CPU 便可从中读取数据和程序。

Cache memory is a very high-speed semiconductor memory which can speed up the CPU. It acts as a buffer between the CPU and main memory. It is used to hold those parts of data and program which are most frequently used by CPU. The parts of data and programs are transferred from the disk to the cache memory by the operating system, from where the CPU can access them.

在本章节，我们将详细阐述高速缓存存储器及其优点和缺点。

In this chapter, we will explain in detail about cache memory along with its advantages and disadvantages.

What is Cache Memory?

在计算机等数字系统中，高速缓存存储器是一种高速易失性半导体存储器，用于存储由系统中央处理器频繁访问的数据和指令。

In digital systems like computers, the cache memory is a high-speed volatile semiconductor memory used to store data and instructions frequently accessed by the central processing unit of the system.

高速缓存存储器充当处理元件和主存储器/主存储器之间的缓冲器，更具体而言是 RAM（随机存取存储器）。它主要用于针对最近和最常使用的数据和程序提供更快的访问。

The cache memory acts as a buffer between the processing element and main/primary memory, more specifically RAM (Random Access Memory). It is mainly used to provide a faster access to the recent and most frequently used data and programs.

高速缓存存储器用于提高数字系统的性能和效率，因为它减少了访问数据所需的时间。

The cache memory is employed for improving the performance and efficiency of the digital systems, as it reduces the time required for accessing the data.

Cache Memory Design

在本文的本节中，我们将探讨缓存存储器设计的不同概念：

In this section of the article, we will discuss different concepts involved in the design of cache memory −

Purpose of the Cache Memory

缓存存储器的主要目的是存储经常使用的数据和指令。这有助于缩短访问时间。

The main purpose of the cache memory is to store frequently used data and instructions. This helps in reducing the access time.

Size of the Cache Memory

已发现，小尺寸高速缓存存储器可显著提升系统的性能。

It is found that small size cache memory results in better performance improvement of the system.

Cache Memory Hierarchy

高速缓存存储器通常按多个层次结构组织，其中每个层次称为高速缓存层次或高速缓存层。计算机系统通常具有多个高速缓存层次，最常见的是 L1（1 级高速缓存）、L2（2 级高速缓存）和 L3（3 级高速缓存）。其中，L1 高速缓存存储器是最小、最快速且最靠近系统 CPU 的，而 L2 和 L3 高速缓存存储器比 L1 高速缓存更大且更慢。

Cache memory is generally organized in multiple hierarchy levels, where each level is called a cache level or cache layer. A computer system typically has multiple cache levels, most common of them are L1 (Level 1 Cache), L2 (Level 2 Cache), and L3 (Level 3 Cache). Here, the cache memory L1 is the smallest, fastest and closest to the CPU of the system, while the L2 and L3 cache memories are larger and slower than L1 cache.

Structure of Cache Memory

高速缓存存储器通常被划分为具有固定大小的块。每个块都具有特定的数据存储容量。高速缓存存储器的结构是通过将所有这些块组合到高速缓存组中而形成的。

Cache memory is typically divided into blocks of a fixed size. Each block has a specific data storage capacity. The structure of the cache memory is formed by grouping all these blocks together into cache sets.

Mapping Techniques for Cache Memory

映射技术用于确定内存块如何映射到高速缓存块。以下三种类型的缓存映射技术通常使用：

The mapping techniques are used to determine how the memory blocks are mapped to cache blocks. The following three types of cache mapping techniques are commonly used −

  1. Direct Mapping − Direct mapping is a simple cache mapping technique in which each memory block is mapped into a certain cache block. Although, this technique can lead to a high rate of conflicts.

  2. Fully Associative Mapping − In this mapping technique, each memory block can be placed in any cache block, hence this technique has high flexibility. However, it requires addition hardware.

  3. Set Associative Mapping − This mapping technique is a combination of direct and fully associative mappings. In this technique, the cache memory is divided into cache sets, and each memory block can be placed in any cache block within its corresponding cache set.

Cache Replacement Algorithms

当一个内存块被要求访问同一个已经被占用的缓存块时，cache replacement algorithm 被用来决定内存块应该被替换成哪个，目的是为新内存块在 cache memory 中腾出空间。

When a memory block is required to be accessed into a cache block that is already occupied, then a cache replacement algorithm is needed to determine which memory block should be replaced to free up space in the cache memory for the new memory block.

以下三个是常用的 cache replacement 算法 −

The following three are the common cache replacement algorithms −

  1. First-In First-Out (FIFO) Algorithm − This algorithm replaces the memory block that exists in the cache memory the longest.

  2. Least Recently Used (LRU) Algorithm − This algorithm replaces the memory block that has been fetched least recently.

  3. Random Replacement (RR) Algorithm − This algorithm replaces any memory block randomly.

Performance of Cache Memory

缓存内存的运行状况通常以命中率来衡量。命中率表明了对内存访问造成 cache memory 命中所占的百分比。命中率较高表明，对内存访问的大部分都利用了 cache memory 来满足需要。这是增强系统运行状况的方法。

The performance of the cache memory is generally measured in terms of its hit rate. The hit rate specifies the percentage of memory accesses that result in cache memory hits. A high hit rate indicates that a significant portion of the memory accesses is satisfied from the cache memory. This provides enhanced system performance.

以上都是缓存内存设计的基本概念。现在，让我们看看 cache memory 设计的优点、缺点。

All these are the fundamental concepts of cache memory design. Now, let’s have look into the advantages and disadvantages of cache memory design.

Types of Cache Memory

按照“层级”对 cache memory 进行分类。其中，每层描述了数字系统中处理元件的可访问性和亲密性。

Cache memory is classified on the basis of "levels". Where, each level describes its accessibility and closeness to the processing element of the digital system.

cache memory 的分类包括以下三个层级 −

The classification of cache memory is done in the following three levels −

L1 (Level 1) Cache Memory

它也被称作 primary cache memory。L1 cache memory 的速度最快。但它体积很小，且主要用在 CPU cache 形式的处理器芯片上。

It is also known as primary cache memory. The L1 cache memory is the fastest one. But it is very small in size and mainly used in the processor chip in the form of CPU cache.

L2 (Level 2) Cache Memory

它也被叫做 secondary cache memory。与 L1 cache memory 相比，它的容量更大。它可以被用在作为 CPU cache 的处理器芯片上，也可以是一个独立的芯片。

It is also called secondary cache memory. It has more capacity as compared to the L1 cache memory. It can be used in the processor chip as CPU cache or it can be a separate chip.

L3 (Level 3) Cache Memory

这是一个专门的 cache memory，设计用于增强 L1 和 L2 cache memory 的运行状况。然而，L3 cache memory 的速度明显低于 L1 或 L2 cache memory。

This one is a specialized cache memory designed to enhance the performance of L1 and L2 cache memories. However, the L3 cache memory is significantly slower than L1 or L2 cache memories.

Features of Cache Memory

cache memory 的关键特点如下 −

The key features of cache memory are listed below −

  1. Cache memory is faster than main memory.

  2. It consumes less access time as compared to main memory.

  3. It stores the program that can be executed within a short period of time.

  4. It stores data for temporary use.

Advantages of Cache Memory

在数字系统中，cache memory 提供了许多的优点，因为它可以改善系统的整体运行状况和效率。cache memory 的一些主要优点如下 −

In digital systems, the cache memory provides several advantages, as it improves the overall performance and efficiency of the system. Some of the key benefits of the cache memory are highlighted below −

  1. Cache memory provides a faster data access speed and reduces the total access time. This characteristic of the cache memory helps to speed up the execution of tasks.

  2. Cache memory helps to reduce the memory latency by storing recent and most frequently used data and instructions. Also, it minimizes the dependency on slower primary memory or RAM. This feature also results in improved system performance and efficiency.

  3. Cache memory operates at the same speed as the CPU. Hence, it can provide a steady stream of input data and instructions that reduces the idle time of the CPU. Therefore, it also improves the CPU utilization.

  4. Cache memory bridges the gap between the high-speed, expensive cache memory and the slow-speed, cheap main memory. It provides a balance between speed, capacity, and cost.

Disadvantages of Cache Memory

然而，cache 内存提供了很多优点。但也有一些劣势，列举如下−

However, the cache memory offers several advantages. But it also has some disadvantages which are listed below −

  1. Cache memory has a very smaller storage capacity. Thus, it cannot be used to hold all the data and instructions required by the processing unit.

  2. Cache memory is expensive to the design and manufacture. It also increases the overall complexity of architecture of the digital system.

  3. Sometimes, the cache pollution may occur when irrelevant data stored in the cache memory and there is no enough space for useful data. This significantly degrades system performance.

Conclusion

总的来说，cache 内存是一种高速半导体内存，主要用于数字系统中以提高其性能和效率。使用 cache 内存可以减少数据访问时间并加速任务执行。但是，作为一种相当昂贵的内存，它会增加系统的整体成本。

In conclusion, the cache memory is a high-speed semiconductor memory primarily used in digital systems to improve their performance and efficiency. The use of cache memory reduces the data access time and speeds up the task execution. However, being a quite expensive memory, it can increase the overall cost of the system.

在本章中，我们涵盖了与 cache 内存相关的所有重要概念，例如其用途、特性、优点和缺点。

In this chapter, we covered all the important concepts related to cache memory such as its purpose, features, advantages, and disadvantages.