---
title: "Global Memory"
canonical: "https://www.thundercompute.com/glossary/memory/global-memory"
description: "The main, large memory space on a GPU accessible by all threads"
sidebarTitle: "Global Memory"
icon: "database"
iconType: "solid"
---

**Global memory** is the largest memory space on a GPU (i.e., VRAM). It is accessible by all threads across all blocks, but has higher latency compared to shared memory or registers.

## Example

```cpp
// Allocate and copy data to global memory
float *d_data;
cudaMalloc(&d_data, n * sizeof(float));
cudaMemcpy(d_data, h_data, n * sizeof(float), cudaMemcpyHostToDevice);
```

## Key Characteristics

- **Size**: Gigabytes (matches VRAM capacity)
- **Latency**: ~400-800 cycles
- **Accessible by**: All threads in all blocks
- **Coalesced access** (consecutive threads reading consecutive addresses) is critical for performance

## See Also

- [VRAM](/gpu-hardware/vram)
- [Shared Memory](/memory/shared-memory)
- [Memory Bandwidth](/memory/memory-bandwidth)