What is the second level of CPU slowdown?

L2cache, also known as L2Cache, is some buffer memory inside the processor that acts just like memory. How did it come about? Back in the 80s, processors were getting faster and faster, and slowly, the speed at which processors needed to read data from memory became higher and higher.

However, the speed of memory is slowly increasing, and memory** that can read and write data at high speeds is too expensive to be used in large quantities. From the perspective of performance, processor design and production companies such as Intel have come up with a way to use a small amount of high-speed memory and a large number of low-speed memory to jointly provide data for the processor. This takes into account the optimal performance and cost of use.

And those high-speed memories are called buffer memory because they are located between the CPU and the memory, and they are the place where the data is stored temporarily. It acts like a temporary storage place in a warehouse, where goods are temporarily stacked in a buffer area when they are lowered from a transport vehicle, and then moved to an internal storage area for long-term storage. The goods are stored in this area for a short period of time, which is a temporary yard.

Initially, there was only one level of cache, but then the processor speed increased, and the first level of cache was not enough, so a second level of cache was added. L2 cache is slower and larger memory than L1 cache, and is mainly used for temporary data exchange between L1 cache and memory.

A piece of static random access memory inside the CPU!

It is used to store intermediate data and results!

His size and performance!It directly affects the performance of the CPU!

The CPU itself is good, fast, but the conversion to memory is discounted, and there has to be a transition, which is caching.

The first-level cache is generally 64k, there is nothing to say, the second-level cache is different, of course, the bigger the better, and you can obviously feel the speed when running large programs.

In fact, the P4 game is expensive because of the high L2 cache.

Level 2 cache is the high-speed memory used by the CPU to store data, you can understand that it is a very high frequency memory, the speed is very block, generally speaking, it is synchronized with the CPU core frequency, the CPU reads the data in order of level 1 data cache Level 2 cache (level 3 cache) memory hard disk (disk), search data from high to low, and search for lower level storage when the data is not in advanced storage.

A Level 2 cache is a higher level of memory than memory and is characterized by a speed block with a small capacity.

First of all, I solemnly declare that the following content is absolutely original, no plagiarism, and I never do that (I have been accused of copying the problem several times after hard work, sad ah......It is hereby declared! ）。

Directly marked as 2M should be shared by all cores inside the CPU to use 2M secondary cache, while 4x256k means that there are four cores inside the CPU, and each core has 256K secondary cache independently.

The difference between these two is that the CPU shares the secondary cache, if a single-threaded task is executed, any core in the CPU can use the shared 2M cache, which can avoid the waste of cache, but there is a problem, if it is not a single-threaded task, so that two cores need to work together, which needs to reasonably allocate the 2M cache to the two cores according to the situation, so that there will be problems, there may be programs that preempt more caches, However, other programs do not have enough cache usage, which causes the program to slow down sharply or even crash.

Having a separate second-level cache for each core avoids the aforementioned uneven distribution of caches when working with multiple threads, as the caches between each core cannot be shared. But conversely, when processing a single-threaded task, that is, when all cores are not required to work at all, the cache of idle cores cannot be used by other cores, which causes a waste of resources.

That's the way to cache these two things, and if you're going to want to go for the highest speed of single-threaded work, you have to give up multi-threaded work, and vice versa. Therefore, when choosing a CPU, you must choose according to your actual needs, not blindly

I studied computer science, the above is a personal opinion, it is not guaranteed to be completely correct, if there is a problem we can communicate with each other, mine.