Knowledge Center
Navigation
Knowledge Center

Central Processing Unit (CPU)

The CPU is an dedicated integrated circuit or IP core that processes logic and math.
popularity

Description

The central processing unit (CPU) is either a dedicated integrated circuit (IC) or intellectual property (IP) core on an IC that processes logic and math. A CPU can handle high-level provisioning of resources for other compute elements more efficiently than other processors, but it is inefficient at multiply accumulate functions.

A CPU IC is often considered the main processor in desktop and laptop computers but also used in embedded systems. Usually designed as for generic uses, CPUs do have variants within families that are designed for specific vertical markets, such as automotive or IoT. The CPU in a computer used to be a system made up of multiple ICs. When the industry was able to put the computer CPU on a single chip, it was called a microprocessor.

CPUs consist of a processing unit called an ALU (arithmetic/logic unit), a control unit that controls how the data comes and out of the chip, registers, operations (fetch, decode, execute) that follow an instruction set architecture (ISA) and input/output. The ISA is described as the boundary between software and hardware. The ISA defines the operands: what type, where stored (which are stored in memory, which are kept on the CPU), and what size.

The CPU has been shrinking in size and scale thanks to Moore’s law, but shrinking the size of a processor is reaching its limits of usefulness.

Multiple CPUs can be installed in a chip as IP cores. Deciding what type of processor to use in an IC means looking at the options, and more than one processor type can be used in the same IC. Clusters of GPUs can process streaming data better than a single GPU, but they are too power-hungry to use everywhere. Clusters of DSPs can do the same for sound, but they’re not very good at classic number crunching. And then there are embedded FPGAs for programmability and security, TPUs for accelerating specific algorithms, and possibly some microcontrollers thrown into the mix. MCUs require less power than a CPU, but a CPU is better for coordinating a variety of functions, such as what gets processed where, what gets prioritized, and how all of this gets orchestrated in time, which is often measured in nanoseconds.

Multimedia

Using GPUs In Semiconductor Manufacturing

Multimedia

Improving Power & Performance Beyond Scaling

Multimedia

High-Performance Memory For AI And HPC

Multimedia

Making Sense Of Inferencing Options

Multimedia

Making Sense Of ML Metrics

Multimedia

Meltdown, Spectre And Foreshadow