XTC, A Research Platform for Optimizing AI Workload Operators
This addresses the need for fair comparison and reuse in AI workload optimization research, though it appears incremental as it builds on existing scheduling concepts with a new interface.
The paper tackles the problem of AI operator optimization being hindered by scheduling languages locked into specific compiler ecosystems, and introduces XTC as a platform that unifies scheduling and performance evaluation across compilers to enable portable experimentation.
Achieving high efficiency on AI operators demands precise control over computation and data movement. However, existing scheduling languages are locked into specific compiler ecosystems, preventing fair comparison, reuse, and evaluation across frameworks. No unified interface currently decouples scheduling specification from code generation and measurement. We introduce XTC, a platform that unifies scheduling and performance evaluation across compilers. With its common API and reproducible measurement framework, XTC enables portable experimentation and accelerates research on optimization strategies.