Inside a modern GPU: from architecture to trillions of calculations

We often speak of modern GPU performance in abstract terms—trillions of operations, ever-increasing computing power. What remains less visible is the structure that makes this possible. A video by Branch Education, using NVIDIA's RTX 3090 GA102 chip as a reference, reveals what happens inside a modern graphics card. At its core is a simple but revealing question: how can a single device perform tens of trillions of calculations per second? The answer lies in their design: unlike CPUs, which process tasks sequentially, GPUs rely on parallel processing, executing thousands of smaller operations simultaneously—an approach known as Single Instruction, Multiple Data (SIMD). This architecture, combined with high-bandwidth memory like GDDR6X, allows GPUs to efficiently handle data-intensive workloads, from rendering to increasingly complex AI tasks. The video shows that performance results from the interaction of these elements within a tightly integrated system, reflecting the expanded role GPUs now play in high-performance computing