MapD: Massive Throughput Database Queries with LLVM on GPUs | NVIDIA Technical BlogMapD leverages NVIDIA GPUs to achieve orders-of-magnitude better performance on standard data analytics tasks by utilizing the massive parallelism and memory bandwidth of GPUs like the NVIDIA Tesla K80 Accelerator.The company uses LLVM-based Just-In-Time (JIT) compilation to transform query plans into architecture-independent intermediate code, allowing for efficient execution on various architect
