|
|
M
e
m
o
r
y
|
|
Memory Guide
The Sharky Extreme Memory Guide deals with issues surrounding memory technology
for the PC, we talk about RAM types, speeds, implications and so forth, dig in:
Part 1 - What is Memory?
The smallest physical unit of memory is the bit, which is essentially an electrical component that is in one of two possible states (usually represented as 0 and 1)...
Part 2 - Memory Performance
The PC architecture is comprised of a few basic pieces: The central processing unit, the memory subsystem, the I/O subsystem, and perhaps a graphics subsystem.
Most discussions about increasing performance focus on the CPU speed, and ignore the others. While it is true that increasing the CPU speed will improve performance to some degree, there is still the problem of getting data to and from main memory...
Part 3 - More on Memory Performance
While processors have been getting faster at a seemingly ever-increasing rate memory speeds have been improving at a much slower rate, mostly due to cost issues. With superscalar architectures and other features being implemented that allow multiple instructions to be executed every clock cycle, this mismatch has become even greater...
Part 4 - Memory Timings
Timing is probably the most interesting and the most difficult to understand aspect of memory operation. For those who wish to get maximum performance from their systems, understanding the various timings is critical.
Though almost all systems in use today utilize SRAM cache, the user generally has little choice over the specific chips used or the controller settings. For this reason, this article will not delve very deeply into SRAM timings, but will instead focus on the various flavors of DRAM...
Part 5 - System Timings
The maximum memory throughput in a system is determined by multiplying the bus speed by the bus width. This means a system with a 64-bit bus running at 100MHz would have a theoretical maximum throughput of 800MB/sec, however it is virtually impossible to achieve this kind of throughput.
One reason it is not possible to actually reach the theoretical maximum is because of the initial latency of the memory. As an example of this, consider an SDRAM module that has 5-1-1-1 timings on a 100MHz bus that is 64 bits wide. Assuming that no other factors slow down the data transfer, this means that in 80ns exactly 32 bytes of data have been transferred. The actual throughput in this case is only 400MB/sec (1 billion / 80ns times 32 bytes). In other words, due to the limitations of the SDRAM itself only half of the bandwidth can be utilized.
Part 6 - Coming Soon
|
|
|