SRDatalog implements worst-case optimal joins on GPUs for Datalog using columnar storage and skew-mitigation techniques, achieving 21-47x speedups on program-analysis workloads while avoiding asymptotic blowups from binary joins.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.DB 2years
2026 2representative citing papers
LLM-synthesized GPU query kernels outperform engineered engines by 7.4x, but a portable SYCL engine with lifted optimizations closes the gap to 1.27x, suggesting engineering remains preferable on GPUs.
citing papers explorer
-
Scaling Worst-Case Optimal Datalog to GPUs
SRDatalog implements worst-case optimal joins on GPUs for Datalog using columnar storage and skew-mitigation techniques, achieving 21-47x speedups on program-analysis workloads while avoiding asymptotic blowups from binary joins.
-
From Custom-Fit to Portable: Bridging the Gap Between Synthesized and Engineered GPU Query Execution
LLM-synthesized GPU query kernels outperform engineered engines by 7.4x, but a portable SYCL engine with lifted optimizations closes the gap to 1.27x, suggesting engineering remains preferable on GPUs.