What is a L2 cache is usually 1MB, or 2MB

The so-called CPU cache is a storage area that comes with the CPU, which is much faster than the memory speed. The so-called L2 cache is slower than the L1 cache, cheaper, but much faster than memory. The low-end ones are more common at 128K and 256K, and the better ones are 512K, and the 1M L2 cache is already very high, and it is now the mainstream high-end chip that AMD and P4 compete in terms of time, such as P4 506 and AMD 64-bit 3000+.

The cache is a part of the CPU, which exists in the CPU.

About L1 and L2 caches.

In order to distinguish between these two concepts, let's first understand RAM.

RAM and ROM are opposite, RAM is the one in which the information disappears after the power is off, and the ROM will not disappear after the power off.

There are two types of RAM, one is static RAM and SRAM; One is dynamic RAM, DRAM. The former is much faster to store than the latter, and the memory we use now is generally dynamic RAM.

Some rookies have said that in order to increase the speed of the system, it is not enough to expand the cache, the larger the expansion, the more data cached, and the faster the system.

The cache is usually static RAM, and the speed is very fast, but the static RAM integration is low (storing the same data, the volume of static RAM is 6 times that of dynamic RAM), **high (static RAM of the same capacity is four times that of dynamic RAM), it can be seen that expanding static RAM as a cache is a very stupid behavior, but in order to improve the performance and speed of the system, we must expand the cache, so that there is a compromise method, not to expand the original static RAM cache, Instead, add some high-speed dynamic RAM as a cache, these high-speed dynamic RAM is faster than regular dynamic RAM, but slower than the original static RAM cache, we call the original static RAM cache Level 1 cache, and the later added dynamic RAM is called the Level 2 cache. The contents of both L1 and L2 caches are replicas (maps) of frequently accessed data in memory, and they both exist to reduce access to slow memory by high-speed CPUs.

Usually, the CPU searches for data or instructions in the order of finding data or instructions: first find it in the first-level cache, find it in the second-level cache, and then find it in the second-level cache, if it can't find it, you can only find it in memory.

It is a moving space that accelerates data exchange, not only 2 m, 4 m, but also 8 m and 16 m.

The bigger, the faster!!

Caches are divided into L1 caches (i.e., L1 caches) and L2 caches (i.e., L2 caches). The CPU first reads data from the L1 cache, then from the L2 cache, and then from memory and virtual memory when it is running, so the capacity and speed of the cache directly affect the working performance of the CPU. L1 caches are built into the CPU and run at the same speed as the CPU, which can effectively improve the efficiency of the CPU.

The larger the L1 cache, the more efficient the CPU will be, but the capacity of the L1 cache is small due to the limitations of the CPU's internal structure. The second-level cache also has a great impact on the CPU operation efficiency, and the current second-level cache is generally integrated in the CPU, but it is divided into two kinds inside the chip, the second-level cache integrated inside the chip is the same frequency as the CPU second-level cache (i.e., full-speed L2 cache), and the operating frequency of the second-level cache integrated outside the chip is half of the operating frequency of the CPU (i.e., half-speed L2 cache), so the operation efficiency is lower.