Graphics Processing Unit
A Graphics Processing Unit (GPU) is a specialized electronic processor designed to accelerate parallel computations, particularly those involving graphics rendering, matrix operations, and large-scale data processing. Unlike a CPU, which prioritizes sequential task execution, a GPU contains thousands of smaller cores optimized for simultaneous processing.
Also Known As
Graphics card processor
Video processor
Parallel processor
Accelerator chip
How It Works
A GPU is built with a massively parallel architecture.
It divides large workloads into thousands of smaller threads that execute concurrently.
Modern GPUs are programmable and support compute frameworks such as NVIDIA CUDA and OpenCL.
They are commonly installed on expansion cards or embedded directly into processors.
Key Characteristics
Thousands of lightweight cores
Optimized for matrix math
High memory bandwidth
Designed for throughput over latency
Common Use Cases
Deep learning training
3D rendering
Video encoding and decoding
GPU vs CPU
| Feature | GPU | CPU |
|---|---|---|
| Core Count | Thousands | 4–64 |
| Best For | Parallel workloads | Sequential logic |
| AI Training | Highly optimized | Limited efficiency |
Benefits
Accelerates AI workloads
Handles massive parallel tasks
Improves rendering performance
Limitations
Not ideal for branching logic
Requires specialized programming
Frequently Asked Questions
What is a GPU used for?
A GPU is used for rendering graphics, accelerating AI training, and processing large parallel workloads.
Is a GPU faster than a CPU?
For parallel tasks like matrix computations, a GPU is significantly faster. For sequential tasks, a CPU performs better.