What is Unified Memory Architecture (UMA)?

The memory reservoir used by the Processor and GPU is the same in UMA, a form of computer memory design. It is frequently used in mobile devices like cellphones and laptops as well as PCs with built-in graphics engines.

For the Processor and GPU, independent graphics processing units have different memory regions. While the GPUs use their own built-in memory libraries of high-speed VRAM for picture data, the Processor stores transient data for apps and functions operating on the computer in the primary system RAM. When required, the PCI Express interface moves data from the main Memory to the VRAM.

The expense and intricacy of the computer are decreased when the CPU has an embedded GPU because there is only one memory space shared by the two processors. Data switching between Memory and VRAM is no longer necessary thanks to this design. Because the CPU and GPU must access memory via a channel that would otherwise only serve the CPU, in some instances this causes efficiency to be delayed.

Apple's M1 line of CPUs marked the debut of their version of a UMA. These processors can reach the common memory pool with much higher speed and much lower delay than other versions because the system's memory is included on the same system-on-a-chip as the CPU and GPU components. Data never needs to leave the processor to go to a different memory section, and neither the CPU nor the GPU are required to transfer data from one cache to another.



You May Interest

What is Block Propagation ?

What is Interoperability in Blockchain Systems ?

What is Sharding in Blockchain ?

What is PKI (Public Key Infrastructure) ?

What are the Fundamental Stages of Data Warehousing ?