Why is server memory more expensive than normal memory?

Both server memory and normal PC memory are memory (RAM), and there is no obvious difference in appearance and structure, but the price is higher than normal memory.

Server memory is also memory, it and we usually see in the computer city in the ordinary PC there is no obvious difference in appearance and structure, it is mainly in the memory of the introduction of some new technology, ordinary PC memory on the server is generally unavailable server can not recognize, This means that the server memory can not be arbitrarily for cheap with ordinary PC memory to replace the reason. Some people have some kind of technology of memory is called "Server memory", in fact, is not comprehensive, the server of these memory technology is currently in the view of the server is dedicated, but can never be guaranteed only server-specific. These new technologies are first used on the server because the server price is more expensive, conditional to apply, these new technologies due to the price of the reasons for the temporary PC machine can not be applied, but with the fall in the price of accessories gradually to the general PC machine, Just like the original parity correction memory was first applied to the server, now is not very common? So server memory is not a specific, it is the application of new memory technology in different time period.

Server memory is mainly SDRAM, DDR and DDR2 three types, the current use of more server memory is DDR and DDR2. The main difference between server memory and normal memory is that the server uses some new technologies, such as ECC (Error checking and correction), Chipkill, register (register), hot-swappable technology, and FB-DIMM (fully buffered memory module). Currently used more server memory mainly using ECC and Reg ECC technology. In terms of appearance, ECC memory has added an ECC-tested particle to meet the need for parity-correcting, and because of the TOSP package, it looks like there are 9 memory particles on each side. REG ECC memory chips are typically 2-3 more than regular motherboards, mainly PLL (Phase Locked Loop) and register ICS. ECC and ECC reg memory from the initial development to now has undergone two generations, the frequency is mainly 133, 266, 333, 400, 533, 667 phase.
(i) Uncover the secret of server memory technology

(1) ecc 
    ECC is an instruction error correction technique widely used in computers in various fields, the English full name of ECC is "error Checking and correcting", and the corresponding Chinese names are called " Error checking and correction, from which we can see that its main function is to "Detect and correct errors", it is more advanced than parity correction technology mainly because it can not only detect errors, but also to correct these errors, these errors after the computer can correctly perform the following tasks, to ensure the normal operation of the server. A lot of time to hear some profiteers said that our server memory is ECC memory, in fact, ECC is not a model, but the application of ECC technology in memory. ECC memory is mainly from SD memory began to be widely used, the current mainstream ECC memory is mainly SD memory.       (2) chipkill 
Chipkill Technology is a new ECC memory protection standard developed by IBM to address the current deficiencies of ECC technology in server memory. ECC memory can only detect and correct single bit errors at the same time, but it is generally powerless to detect errors in more than two bits of data at the same time. Since the server currently in use has a high system speed, and the phenomenon of multiple bit error seldom occurs, ECC technology is fully recognized and applied, making ECC memory technology become the memory standard on almost all servers. &NBSP
with the CPU performance of an intel-based processor-based server increasing at a geometric level, the performance of the hard drive does not keep up with CPU performance, so the server needs a lot of memory to temporarily save the data that needs to be read on the CPU. Such large data traffic results in a single memory chip that typically provides 4 (32-bit) or 8 (64-bit) bits of data per visit, read so much data at once, the likelihood of multiple-bit data errors will be greatly improved, and ECC can not correct the errors above the double bit, This is likely to result in the loss of all bit data, and the system crashes quickly. IBM's Chipkill technology is the use of memory substructure method to solve this problem. A single chip, regardless of the width of the data, only for a given ECC identifier, it has the most impact of one bit. Therefore, even if the entire memory chip out of the fault, each ECC identification code will appear up to a bit of bad data, and this situation can be fixed through ECC logic, so as to ensure the fault tolerance of the memory subsystem, the server in the event of failure, there is a strong ability to recover. With this memory technology, memory can simultaneously check and repair 4 error data bits, the server reliability and stability is more fully protected. (3) register 
   register is a register or directory register, the function of memory as the directory of books, when the memory is read and write instructions, will first retrieve this directory, and then read and write operations, which will greatly improve the server memory efficiency. Memory with register must have buffer (buffer), and the current can see the

Intel Big study opens one-stop retrieval times large data
Cloud computing (Cloud)
Mobility (adroit)
Other (others)  
 wk_ad_begin ({pid:21 Wk_ad_after, Function () {$ ('. Ad-hidden '). Hide ();}, function () {$ ('. Ad-hidden '). Show ();});   &NBSP
  
Register memory also has ECC capabilities, which are primarily applied to high-end servers and graphics workstations, such as IBM Netfinity 5000. (4) FB-DIMM&NBSP
Full buffer memory Module FB-DIMM (fully BUFFERED-DIMM) is a new memory module and interconnect architecture developed by Intel on DDR2 and DDR3, which can match the current DDR2 memory chip. It can also be paired with future DDR3 memory chips. Fb-dimm can greatly increase the system memory bandwidth and greatly maximize memory capacity. Compared with DDR2 memory, FB-DIMM can provide four times times more bandwidth than normal memory at the same memory frequency, and can support up to 24 times times the maximum memory capacity, and the system can support 192GB memory. Compared with the common DIMM module technology, the data and command transmission between FB-DIMM and memory controller is no longer a traditional parallel line, but the design of parallel serial interface is similar to that of pci-express, which is transmitted in serial way. In this new architecture, the buffers on each DIMM are concatenated, the point-to-point connection, and the data is passed down a buffer after the first buffer so that the connection impedance between the first buffer and the memory controller is always stable, thus contributing to increased capacity and frequency. &NBSP