What is it about?
Processing database operations on different hardware devices (GPU, FPGA, CPU) requieres expert knowledge and specific implementation for each architecture. We aim massively simplify using heterogeneous hardware by introducing a vector based abstraction layer.
Featured Image
Why is it important?
The effort to maintain code bases across multiple architectures is prohibitivly expensive. By introducing a vector based abstraction layer (Template Vector Library), programming against different architectures is much simpler. The same code runs on different devices.
Read the Original
This page is a summary of: The Case for SIMDified Analytical Query Processing on GPUs, June 2021, ACM (Association for Computing Machinery),
DOI: 10.1145/3465998.3466015.
You can read the full text:
Contributors
The following have contributed to this page
