CPU Socket
What is it?
Also known as a CPU slot, the CPU socket is used to connect a microprocessor with a printed circuit board (PCB) without soldering, allowing for the CPU to be installed or replaced more easily on the motherboard. Commonly found in PCs and servers, CPU sockets are the antithesis of the surface-mount CPUs used in laptops, which conserve space at the expense of convenient installation or swapping.
Common types of CPU sockets include the Pin Grid Array (PGA) and the Land Grid Array (LGA). The difference between the two is that PGA places the pins on the processor and the holes in the socket, whereas LGA has a socket with pins that you place the processor on. The number and arrangement of pins vary according to the types of CPUs used, but generally speaking, greater pin count and density will allow more data signals to be transferred more quickly.
Common types of CPU sockets include the Pin Grid Array (PGA) and the Land Grid Array (LGA). The difference between the two is that PGA places the pins on the processor and the holes in the socket, whereas LGA has a socket with pins that you place the processor on. The number and arrangement of pins vary according to the types of CPUs used, but generally speaking, greater pin count and density will allow more data signals to be transferred more quickly.
Why do you need it?
Because a CPU socket connects the brains of the server—the CPU—to everything else, there is no way to work without a CPU socket. Also, since CPU sockets may appear physically identical but are in fact electrically incompatible, you should not swap processors just because they use the same socket-type—always make sure to match your CPU sockets with the CPUs they were designed for.
How is GIGABYTE helpful?
GIGABYTE servers support various different processors, including Intel Xeon, AMD EPYC, and Ampere Altra built with ARM architecture. Each server is outfitted with CPU sockets that match their processors. Depending on the type of server, there may be just one socket, or multiple sockets for tasks such as parallel computing or heterogeneous computing.
