RAM or random access memory is an incredibly important part of any modern computer. The CPU (central processing unit) of a computer needs data and instructions to do the job. This information needs to be stored somewhere. Somewhere is called computer memory.
There are different types of RAM, each with their own pros and cons. CPUs have a very small amount of built-in memory, known as CPU “cache”. This memory is incredibly fast and is essentially part of the processor itself. However, it is very expensive and cannot be used as the main memory of a computer.
This is where RAM comes into play. Random access memory is presented in the form of silicon computer chips connected to the memory bus. Cache memory on the CPU itself is actually also a form of RAM, but when the term is commonly used, it refers to memory chips that are outside the CPU.
A memory bus is simply a special set of circuits that move information between the CPU and the RAM itself. The operating system moves information off the system’s much slower mechanical or solid state hard drive in preparation for the needs of the CPU. For example, when a video game is “loaded,” data is moved from the hard drive to RAM.
As an analogy, think of RAM as the top of your desk and drawers as your hard drive, with you playing the role of the CPU. Working with objects on the table is quick and easy, but there is not much space. This means you need to move items between the tabletop and drawers as needed.
Computers, smartphones, game consoles and all other computing devices in use today have some type of RAM We’ll go over each of them, explaining how they work and what they are used for. Specifically, we’ll look at the following types of RAM:
– /
- SRAM
- DRAM
- SDRAM
- SDR RAM
- DDR SDRAM
- GDDR
- HMB
Don’t worry if this sounds scary gibberish. Everything will be clear soon.
SRAM – Fixed RAM
One of the two main types of RAM, SRAM, is special because it doesn’t need to be “updated” to retain the information it currently stores. As long as power flows through the circuits, information remains where it is.
SRAM consists of several transistors (4-6) and is incredibly fast due to its nature. It is relatively difficult and expensive, however, which is why you will find it in processors used as ultra-fast caches.
There are also small amounts of SRAM cache where data needs to move quickly but can be bottlenecks. Hard disk buffers are a good example of this use case. Wherever the device has more data, chances are there will be some sort of SRAM to help smooth out that transfer.
DRAM – dynamic random access memory
DRAM is another common type of RAM design. DRAM is built using transistors and capacitors. If you don’t update every memory location, it will lose its contents. This is why it is called “dynamic” and not “static”.
DRAM is much slower than SRAM, but still much faster than secondary storage devices like hard drives. It is also much cheaper than SRAM, and it is typical for computers to have several gigabytes of DRAM on board as their primary RAM solution.
SDRAM – Concurrent Dynamic Random Access Memory
Some people think SDRAM is a mixture of SRAM and DRAM, but it is not! This is DRAM that has been synchronized with the processor clock speed.
The DRAM will wait for the CPU before responding to data input requests. Due to its synchronous nature and the way SDRAM is configured in banks, the CPU can execute multiple instructions at the same time, greatly increasing its overall performance.
SDRAM is the main form of the basic type of RAM used in most computers today. It is also known as SDR SDRAM or single bit rate synchronous dynamic random access memory . While this is essentially the same type of memory used in computers today, its standard form SDR is largely outdated and replaced by the next type of RAM on our list.
Double Data Rate Concurrent RAM
The first thing you should know is that there are several generations of DDR memory. The first generation, which we call DDR 1 in retrospect, doubled the speed of SDRAM, allowing read and write operations to occur at both the peak and minimum clock cycle.
DDR2, DDR3 and DDR4 today have improved significantly over the first generation DDR. The performance of these memory modules is measured in megatransfers per second or “MT / S”. One mega-transfer is equivalent to one million clock cycles. The fastest first-generation DDR chips could run at 400 MT / s. DDR4 can go up to 3200 MT / s!
GDDR SDRAM – Dual Graphics Data Rate RAM
GDDR is currently in its sixth generation and almost always connects to the GPU (GPU) on a video card or game console. GDDR is associated with regular DDR, but is intended for graphics use cases. Emphasizing huge bandwidth, worrying less about low latency.
In other words, this memory does not respond as quickly as regular SDRAM, but it can move more information at once when it does respond. This is ideal for graphics applications where many gigabytes of texture data need to be transferred to render a scene and the low latency has no real impact.
Despite its name, GDDR can be used just like regular system RAM. For example, the PlayStation 4 has a single pool of GDDR memory that developers can share in any way they want, distributing parts between the CPU and GPU if necessary.
HBM – High Bandwidth Memory
GDDR has a competitor in the form of HBM memory, which is present on a limited number of AMD graphics cards. Currently, the latest version is HBM 2, but it is unknown if it will replace GDDR or cease to exist.
The most important part of memory performance is the total amount of data that can be shifted in a given amount of time. One way to do this is to make memory very fast. Another way to improve overall throughput is to widen the data “pipe”.
HBM runs at lower raw clock speeds than GDDR, but uses a unique 3D stack chip design that provides a very wide physical path for data as well as much shorter signal distances. The end result is a memory solution with the same overall bandwidth as GDDR, but with lower latency.
The problem with HBM is that it is difficult to manufacture, and due to its physical design, it is not yet possible to achieve the capacity that is trivial for GDDR. If these problems are eventually overcome, it could replace the GDDR, but there is no guarantee that this will happen.
Thanks for the memories!
It should be obvious that RAM is an essential component of any computer, and when it fails, it can be difficult to figure out what the problem really is.
After all, an attacker here or there could make your system slightly unstable or cause seemingly random crashes. This is why you should always check for bad RAM when you have unexplained stability issues.
We may be able to move beyond RAM someday, but this is going to be an important piece of the computing performance puzzle for the foreseeable future, so we might as well learn about it.
–