cd /news/large-language-models/datakernelbench-can-llms-optimize-da… · home topics large-language-models article
[ARTICLE · art-112658] src=arxiv.org ↗ pub= topic=large-language-models verified=true sentiment=· neutral

DataKernelBench: Can LLMs Optimize Database Queries on GPUs?

A new benchmark, DataKernelBench, evaluates whether large language models can optimize database queries on GPUs, achieving up to 2.11x speedup over torch.compile on TPC-H SF10 with an H100 GPU. The benchmark translates SQL into PyTorch TorchPlan programs and tests ten proprietary and open-weight models, finding that stronger models benefit most from full-query specialization and that kernel fusion is key. On TPC-H SF100 with four H100 GPUs, the approach achieves 2.54x speedup using Dask-cuDF for on-demand partition loading.

read1 min views1 publishedAug 27, 2026

arXiv:2608.25061v1 Announce Type: new Abstract: GPUs increasingly accelerate database systems, but query-specific peak performance still often relies on hand-written kernels. Existing LLM kernel benchmarks focus on machine learning operators, leaving irregular, heterogeneous, data-movement-heavy database-style operators untested. We introduce DataKernelBench, which translates SQL into validated PyTorch TorchPlan programs and evaluates LLMs that optimize either the core tensor-bounded snippet or the full query in CUDA or Triton through execution-guided repair. Across ten proprietary and open-weight models on TPC-H SF10 with an H100 GPU, the strongest full-query CUDA configuration achieves $2.11\times$ speedup over torch.compile at full pass rate. We find that higher-performing implementations commonly use kernel fusion and execution-strategy changes, stronger models benefit most from full-query specialization, and workload context matters more than hardware context. To handle data larger than GPU memory, we extend TorchPlan with Dask-cuDF for on-demand partition on TPC-H SF100 with four H100 GPUs, achieving $2.54\times$ speedup

── more in #large-language-models 4 stories · sorted by recency
── more on @datakernelbench 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/datakernelbench-can-…] indexed:0 read:1min 2026-08-27 ·