First, the level of understanding of the structure of the memory
1, and read books to be in control
1, CPU metaphor of the computer's "brain" of the things we think about, like the CPU register
2, and our brain's memory, like CPU Cache
2, the SRAM
1, SRAM Why is called static memory
2, is not high storage density SRAM
3, however, because the SRAM circuit is simple, so the access speed is very fast
3, three cache CPU's
1、L1
2, L2
3, L3
4, and comparative reading
4、DRAM
1, and SRAM comparison
2, memory type
3, a larger storage capacity
4, how to maintain the data is not lost
Second, the memory hierarchy of
1, the nature of
2, brain (CPU) in the memory (L1 Cache)
3, our own study desk and brain
4, CPU, and does not deal directly with each of the memory device
Third, now the mainstream notebook computer analysis
When using a memory, how to weigh the price and performance?
Performance differences and price differences between different levels of memory, are at least an order of magnitude. L1 Cache access time delay of 1 nanosecond (ns), and the memory is already a 100 nanoseconds. In terms of price, both of which are poor out of 400 times.
I am here to put a variety of memory cost comparison table, you can look at. You can also click on this link, by dragging the view from 1990 to 2020 with the progress of hardware, changes in access latency.
Because of this difference in price and performance, you will see that our actual computer hardware configuration is performed when the portfolio allocation will go to a variety of storage devices.
We can now find a mainstream laptop to look at, for example, an entry-level HP war 66 laptop. Today, the price of Jingdong is 4999 yuan. Its configuration is described below.
- 1. Intle i5-8265U a CPU (which is a 4-core CPU)
- 2 ,. 8GB of memory
- 3. a 128G SSD drive
- A hard disk in the HDD 4. 1T
You can see that in a real computer inside, the more quickly the device, the smaller capacity. A total of more than ten trillion Cache here, costs only tens of dollars. The 8GB of RAM, SSD 128G and 1T of HDD, probably retail price together,
And we will cache the price almost.
IV Summary extension
这节的内容不知道你掌握了多少呢？为了帮助你记忆，我这里再带你复习一下本节的重点。
我们常常把CPU比喻成高速运转的大脑，那么和大脑同步的寄存器（Register），就存放着我们当下正在思考和处理的数据。而L1-L3的CPU Cache，
好比存放在我们大脑中的短期到长期的记忆。我们需要小小花费一点时间，就能调取并进行处理。
我们自己的书桌书架就好比计算机的内存，能放下更多的书也就是数据，但是找起来和看起来就要慢上不少。而图书馆更像硬盘这个外存，能够放下更多的数据，
找起来也更费时间。从寄存器、CPU Cache，到内存、硬盘，这样一层层下来的存储器，速度越来越慢，空间越来越大，价格也越来越便宜。
这三个“越来越”的特性，使得我们在组装计算机的时候，要组合使用各种存储设备。越是快且贵的设备，实际在一台计算机里面的存储空间往往就越小。
而越是慢且便宜的设备，在实际组装的计算机里面的存储空间就会越大。
在后面的关于存储器的内容里，我会带着你进一步深入了解，各个层次的存储器是如何运作的，在不同类型的应用和性能要求下，是否可以靠人工添加一层缓存层来解决问题，
以及在程序开发层面，如何利用好不同层次的存储器设备的访问原理和特性。